Ubuntu
commited on
Commit
·
7820b0f
0
Parent(s):
added models
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +74 -0
- app.py +15 -0
- requirements.txt +9 -0
- research/ai_detector/checkpoint-281/config.json +63 -0
- research/ai_detector/checkpoint-281/model.safetensors +3 -0
- research/ai_detector/checkpoint-281/optimizer.pt +3 -0
- research/ai_detector/checkpoint-281/preprocessor_config.json +22 -0
- research/ai_detector/checkpoint-281/rng_state.pth +0 -0
- research/ai_detector/checkpoint-281/scheduler.pt +0 -0
- research/ai_detector/checkpoint-281/trainer_state.json +198 -0
- research/ai_detector/checkpoint-281/training_args.bin +0 -0
- research/ai_detector/checkpoint-562/config.json +63 -0
- research/ai_detector/checkpoint-562/model.safetensors +3 -0
- research/ai_detector/checkpoint-562/optimizer.pt +3 -0
- research/ai_detector/checkpoint-562/preprocessor_config.json +22 -0
- research/ai_detector/checkpoint-562/rng_state.pth +0 -0
- research/ai_detector/checkpoint-562/scheduler.pt +0 -0
- research/ai_detector/checkpoint-562/trainer_state.json +375 -0
- research/ai_detector/checkpoint-562/training_args.bin +0 -0
- research/ai_detector/checkpoint-843/config.json +63 -0
- research/ai_detector/checkpoint-843/model.safetensors +3 -0
- research/ai_detector/checkpoint-843/optimizer.pt +3 -0
- research/ai_detector/checkpoint-843/preprocessor_config.json +22 -0
- research/ai_detector/checkpoint-843/rng_state.pth +0 -0
- research/ai_detector/checkpoint-843/scheduler.pt +0 -0
- research/ai_detector/checkpoint-843/trainer_state.json +552 -0
- research/ai_detector/checkpoint-843/training_args.bin +0 -0
- research/ai_detector_v2/checkpoint-1210/config.json +63 -0
- research/ai_detector_v2/checkpoint-1210/model.safetensors +3 -0
- research/ai_detector_v2/checkpoint-1210/optimizer.pt +3 -0
- research/ai_detector_v2/checkpoint-1210/preprocessor_config.json +22 -0
- research/ai_detector_v2/checkpoint-1210/rng_state.pth +0 -0
- research/ai_detector_v2/checkpoint-1210/scheduler.pt +0 -0
- research/ai_detector_v2/checkpoint-1210/trainer_state.json +765 -0
- research/ai_detector_v2/checkpoint-1210/training_args.bin +0 -0
- research/ai_detector_v2/checkpoint-1815/config.json +63 -0
- research/ai_detector_v2/checkpoint-1815/model.safetensors +3 -0
- research/ai_detector_v2/checkpoint-1815/optimizer.pt +3 -0
- research/ai_detector_v2/checkpoint-1815/preprocessor_config.json +22 -0
- research/ai_detector_v2/checkpoint-1815/rng_state.pth +0 -0
- research/ai_detector_v2/checkpoint-1815/scheduler.pt +0 -0
- research/ai_detector_v2/checkpoint-1815/trainer_state.json +1134 -0
- research/ai_detector_v2/checkpoint-1815/training_args.bin +0 -0
- research/ai_detector_v2/checkpoint-605/config.json +63 -0
- research/ai_detector_v2/checkpoint-605/model.safetensors +3 -0
- research/ai_detector_v2/checkpoint-605/optimizer.pt +3 -0
- research/ai_detector_v2/checkpoint-605/preprocessor_config.json +22 -0
- research/ai_detector_v2/checkpoint-605/rng_state.pth +0 -0
- research/ai_detector_v2/checkpoint-605/scheduler.pt +0 -0
- research/ai_detector_v2/checkpoint-605/trainer_state.json +390 -0
.gitattributes
ADDED
|
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
/research/ai_detector/checkpoint-281/model.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
/research/ai_detector/checkpoint-281/optimizer.pt filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
/research/ai_detector/checkpoint-562/model.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
/research/ai_detector/checkpoint-562/optimizer.pt filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
/research/ai_detector/checkpoint-843/model.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
/research/ai_detector/checkpoint-843/optimizer.pt filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
/research/ai_detector_v2/checkpoint-1210/model.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
/research/ai_detector_v2/checkpoint-1210/optimizer.pt filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
/research/ai_detector_v2/checkpoint-1815/model.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
/research/ai_detector_v2/checkpoint-1815/optimizer.pt filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
/research/ai_detector_v2/checkpoint-605/model.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
/research/ai_detector_v2/checkpoint-605/optimizer.pt filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
/research/dataset/test/data-00000-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
/research/dataset/test/data-00001-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
/research/dataset/test/data-00002-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
/research/dataset/test/data-00003-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
/research/dataset/test/data-00004-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
/research/dataset/test/data-00005-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
/research/dataset/test/data-00006-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
/research/dataset/test/data-00007-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
/research/dataset/test/data-00008-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
/research/dataset/test/data-00009-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
/research/dataset/test/data-00010-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
/research/dataset/test/data-00011-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
/research/dataset/test/data-00012-of-00013.arrow filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
/research/dataset/train/data-00000-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
/research/dataset/train/data-00001-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
/research/dataset/train/data-00002-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
/research/dataset/train/data-00003-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
/research/dataset/train/data-00004-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
/research/dataset/train/data-00005-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
/research/dataset/train/data-00006-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
/research/dataset/train/data-00007-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
/research/dataset/train/data-00008-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
/research/dataset/train/data-00009-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
/research/dataset/train/data-00010-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
/research/dataset/train/data-00011-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
/research/dataset/train/data-00012-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
/research/dataset/train/data-00013-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
/research/dataset/train/data-00014-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
/research/dataset/train/data-00015-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
/research/dataset/train/data-00016-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
/research/dataset/train/data-00017-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
/research/dataset/train/data-00018-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
/research/dataset/train/data-00019-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
/research/dataset/train/data-00020-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
/research/dataset/train/data-00021-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 48 |
+
/research/dataset/train/data-00022-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 49 |
+
/research/dataset/train/data-00023-of-00024.arrow filter=lfs diff=lfs merge=lfs -text
|
| 50 |
+
/research/dataset/validation/data-00000-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 51 |
+
/research/dataset/validation/data-00001-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 52 |
+
/research/dataset/validation/data-00002-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 53 |
+
/research/dataset/validation/data-00003-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 54 |
+
/research/dataset/validation/data-00004-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 55 |
+
/research/dataset/validation/data-00005-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 56 |
+
/research/dataset/validation/data-00006-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 57 |
+
/research/dataset/validation/data-00007-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 58 |
+
/research/dataset/validation/data-00008-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 59 |
+
/research/dataset/validation/data-00009-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 60 |
+
/research/dataset/validation/data-00010-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 61 |
+
/research/dataset/validation/data-00011-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 62 |
+
/research/dataset/validation/data-00012-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 63 |
+
/research/dataset/validation/data-00013-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 64 |
+
/research/dataset/validation/data-00014-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 65 |
+
/research/dataset/validation/data-00015-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 66 |
+
/research/dataset/validation/data-00016-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 67 |
+
/research/dataset/validation/data-00017-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 68 |
+
/research/dataset/validation/data-00018-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 69 |
+
/research/dataset/validation/data-00019-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 70 |
+
/research/dataset/validation/data-00020-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 71 |
+
/research/dataset/validation/data-00021-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 72 |
+
/research/dataset/validation/data-00022-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 73 |
+
/research/dataset/validation/data-00023-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
| 74 |
+
/research/dataset/validation/data-00024-of-00025.arrow filter=lfs diff=lfs merge=lfs -text
|
app.py
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from transformers import pipeline
|
| 2 |
+
from PIL import Image
|
| 3 |
+
import streamlit as st
|
| 4 |
+
classifier = pipeline("image-classification", model="research/ai_detector_v2/checkpoint-1815")
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
with st.sidebar:
|
| 8 |
+
image= st.sidebar.file_uploader("Input image:", type=["png", "jpg"])
|
| 9 |
+
btn= st.button("Submit")
|
| 10 |
+
|
| 11 |
+
if btn:
|
| 12 |
+
img= original_image= Image.open(image)
|
| 13 |
+
st.json(
|
| 14 |
+
classifier(img)
|
| 15 |
+
)
|
requirements.txt
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
transformers
|
| 2 |
+
evaluate
|
| 3 |
+
datasets
|
| 4 |
+
Pillow
|
| 5 |
+
torch
|
| 6 |
+
torchvision
|
| 7 |
+
scikit-learn
|
| 8 |
+
accelerate
|
| 9 |
+
streamlit
|
research/ai_detector/checkpoint-281/config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "umm-maybe/AI-image-detector",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SwinForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"depths": [
|
| 8 |
+
2,
|
| 9 |
+
2,
|
| 10 |
+
18,
|
| 11 |
+
2
|
| 12 |
+
],
|
| 13 |
+
"drop_path_rate": 0.1,
|
| 14 |
+
"embed_dim": 128,
|
| 15 |
+
"encoder_stride": 32,
|
| 16 |
+
"hidden_act": "gelu",
|
| 17 |
+
"hidden_dropout_prob": 0.0,
|
| 18 |
+
"hidden_size": 1024,
|
| 19 |
+
"id2label": {
|
| 20 |
+
"0": "ai_gen",
|
| 21 |
+
"1": "human"
|
| 22 |
+
},
|
| 23 |
+
"image_size": 224,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"label2id": {
|
| 26 |
+
"ai_gen": "0",
|
| 27 |
+
"human": "1"
|
| 28 |
+
},
|
| 29 |
+
"layer_norm_eps": 1e-05,
|
| 30 |
+
"max_length": 128,
|
| 31 |
+
"mlp_ratio": 4.0,
|
| 32 |
+
"model_type": "swin",
|
| 33 |
+
"num_channels": 3,
|
| 34 |
+
"num_heads": [
|
| 35 |
+
4,
|
| 36 |
+
8,
|
| 37 |
+
16,
|
| 38 |
+
32
|
| 39 |
+
],
|
| 40 |
+
"num_layers": 4,
|
| 41 |
+
"out_features": [
|
| 42 |
+
"stage4"
|
| 43 |
+
],
|
| 44 |
+
"out_indices": [
|
| 45 |
+
4
|
| 46 |
+
],
|
| 47 |
+
"padding": "max_length",
|
| 48 |
+
"patch_size": 4,
|
| 49 |
+
"path_norm": true,
|
| 50 |
+
"problem_type": "single_label_classification",
|
| 51 |
+
"qkv_bias": true,
|
| 52 |
+
"stage_names": [
|
| 53 |
+
"stem",
|
| 54 |
+
"stage1",
|
| 55 |
+
"stage2",
|
| 56 |
+
"stage3",
|
| 57 |
+
"stage4"
|
| 58 |
+
],
|
| 59 |
+
"torch_dtype": "float32",
|
| 60 |
+
"transformers_version": "4.37.2",
|
| 61 |
+
"use_absolute_embeddings": false,
|
| 62 |
+
"window_size": 7
|
| 63 |
+
}
|
research/ai_detector/checkpoint-281/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:15a1521fef8e655b49176248ecc2bd0a275b3b7c2e4d3ff372f55a6afbe5f1ec
|
| 3 |
+
size 347498816
|
research/ai_detector/checkpoint-281/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ca28a5528de829c1aa0c706f308e3a9084c481b474a7bb2983dfc59db461795
|
| 3 |
+
size 694317645
|
research/ai_detector/checkpoint-281/preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
research/ai_detector/checkpoint-281/rng_state.pth
ADDED
|
Binary file (14.2 kB). View file
|
|
|
research/ai_detector/checkpoint-281/scheduler.pt
ADDED
|
Binary file (1.06 kB). View file
|
|
|
research/ai_detector/checkpoint-281/trainer_state.json
ADDED
|
@@ -0,0 +1,198 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9964804552942939,
|
| 3 |
+
"best_model_checkpoint": "ai_detector/checkpoint-281",
|
| 4 |
+
"epoch": 0.9991111111111111,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 281,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"learning_rate": 5.882352941176471e-06,
|
| 14 |
+
"loss": 0.6876,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.07,
|
| 19 |
+
"learning_rate": 1.1764705882352942e-05,
|
| 20 |
+
"loss": 0.4009,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.11,
|
| 25 |
+
"learning_rate": 1.7647058823529414e-05,
|
| 26 |
+
"loss": 0.2111,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.14,
|
| 31 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 32 |
+
"loss": 0.0845,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.18,
|
| 37 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 38 |
+
"loss": 0.052,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.21,
|
| 43 |
+
"learning_rate": 3.529411764705883e-05,
|
| 44 |
+
"loss": 0.0831,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.25,
|
| 49 |
+
"learning_rate": 4.11764705882353e-05,
|
| 50 |
+
"loss": 0.0719,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.28,
|
| 55 |
+
"learning_rate": 4.705882352941177e-05,
|
| 56 |
+
"loss": 0.0493,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.32,
|
| 61 |
+
"learning_rate": 4.967018469656992e-05,
|
| 62 |
+
"loss": 0.0278,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.36,
|
| 67 |
+
"learning_rate": 4.901055408970976e-05,
|
| 68 |
+
"loss": 0.0447,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.39,
|
| 73 |
+
"learning_rate": 4.835092348284961e-05,
|
| 74 |
+
"loss": 0.0597,
|
| 75 |
+
"step": 110
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.43,
|
| 79 |
+
"learning_rate": 4.7691292875989446e-05,
|
| 80 |
+
"loss": 0.0194,
|
| 81 |
+
"step": 120
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.46,
|
| 85 |
+
"learning_rate": 4.703166226912929e-05,
|
| 86 |
+
"loss": 0.0308,
|
| 87 |
+
"step": 130
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.5,
|
| 91 |
+
"learning_rate": 4.6372031662269136e-05,
|
| 92 |
+
"loss": 0.0328,
|
| 93 |
+
"step": 140
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.53,
|
| 97 |
+
"learning_rate": 4.5712401055408974e-05,
|
| 98 |
+
"loss": 0.0415,
|
| 99 |
+
"step": 150
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.57,
|
| 103 |
+
"learning_rate": 4.505277044854881e-05,
|
| 104 |
+
"loss": 0.0075,
|
| 105 |
+
"step": 160
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.6,
|
| 109 |
+
"learning_rate": 4.439313984168866e-05,
|
| 110 |
+
"loss": 0.0358,
|
| 111 |
+
"step": 170
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.64,
|
| 115 |
+
"learning_rate": 4.3733509234828496e-05,
|
| 116 |
+
"loss": 0.0684,
|
| 117 |
+
"step": 180
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.68,
|
| 121 |
+
"learning_rate": 4.307387862796834e-05,
|
| 122 |
+
"loss": 0.0382,
|
| 123 |
+
"step": 190
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.71,
|
| 127 |
+
"learning_rate": 4.2414248021108186e-05,
|
| 128 |
+
"loss": 0.0252,
|
| 129 |
+
"step": 200
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.75,
|
| 133 |
+
"learning_rate": 4.1754617414248024e-05,
|
| 134 |
+
"loss": 0.0356,
|
| 135 |
+
"step": 210
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.78,
|
| 139 |
+
"learning_rate": 4.109498680738786e-05,
|
| 140 |
+
"loss": 0.0214,
|
| 141 |
+
"step": 220
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.82,
|
| 145 |
+
"learning_rate": 4.043535620052771e-05,
|
| 146 |
+
"loss": 0.0137,
|
| 147 |
+
"step": 230
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.85,
|
| 151 |
+
"learning_rate": 3.9775725593667545e-05,
|
| 152 |
+
"loss": 0.091,
|
| 153 |
+
"step": 240
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.89,
|
| 157 |
+
"learning_rate": 3.911609498680739e-05,
|
| 158 |
+
"loss": 0.0103,
|
| 159 |
+
"step": 250
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.92,
|
| 163 |
+
"learning_rate": 3.8456464379947235e-05,
|
| 164 |
+
"loss": 0.0418,
|
| 165 |
+
"step": 260
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.96,
|
| 169 |
+
"learning_rate": 3.779683377308707e-05,
|
| 170 |
+
"loss": 0.0089,
|
| 171 |
+
"step": 270
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.0,
|
| 175 |
+
"learning_rate": 3.713720316622691e-05,
|
| 176 |
+
"loss": 0.0347,
|
| 177 |
+
"step": 280
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.0,
|
| 181 |
+
"eval_accuracy": 0.9964804552942939,
|
| 182 |
+
"eval_loss": 0.013468941673636436,
|
| 183 |
+
"eval_runtime": 252.3562,
|
| 184 |
+
"eval_samples_per_second": 52.917,
|
| 185 |
+
"eval_steps_per_second": 3.309,
|
| 186 |
+
"step": 281
|
| 187 |
+
}
|
| 188 |
+
],
|
| 189 |
+
"logging_steps": 10,
|
| 190 |
+
"max_steps": 843,
|
| 191 |
+
"num_input_tokens_seen": 0,
|
| 192 |
+
"num_train_epochs": 3,
|
| 193 |
+
"save_steps": 500,
|
| 194 |
+
"total_flos": 1.410220001304576e+18,
|
| 195 |
+
"train_batch_size": 16,
|
| 196 |
+
"trial_name": null,
|
| 197 |
+
"trial_params": null
|
| 198 |
+
}
|
research/ai_detector/checkpoint-281/training_args.bin
ADDED
|
Binary file (4.66 kB). View file
|
|
|
research/ai_detector/checkpoint-562/config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "umm-maybe/AI-image-detector",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SwinForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"depths": [
|
| 8 |
+
2,
|
| 9 |
+
2,
|
| 10 |
+
18,
|
| 11 |
+
2
|
| 12 |
+
],
|
| 13 |
+
"drop_path_rate": 0.1,
|
| 14 |
+
"embed_dim": 128,
|
| 15 |
+
"encoder_stride": 32,
|
| 16 |
+
"hidden_act": "gelu",
|
| 17 |
+
"hidden_dropout_prob": 0.0,
|
| 18 |
+
"hidden_size": 1024,
|
| 19 |
+
"id2label": {
|
| 20 |
+
"0": "ai_gen",
|
| 21 |
+
"1": "human"
|
| 22 |
+
},
|
| 23 |
+
"image_size": 224,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"label2id": {
|
| 26 |
+
"ai_gen": "0",
|
| 27 |
+
"human": "1"
|
| 28 |
+
},
|
| 29 |
+
"layer_norm_eps": 1e-05,
|
| 30 |
+
"max_length": 128,
|
| 31 |
+
"mlp_ratio": 4.0,
|
| 32 |
+
"model_type": "swin",
|
| 33 |
+
"num_channels": 3,
|
| 34 |
+
"num_heads": [
|
| 35 |
+
4,
|
| 36 |
+
8,
|
| 37 |
+
16,
|
| 38 |
+
32
|
| 39 |
+
],
|
| 40 |
+
"num_layers": 4,
|
| 41 |
+
"out_features": [
|
| 42 |
+
"stage4"
|
| 43 |
+
],
|
| 44 |
+
"out_indices": [
|
| 45 |
+
4
|
| 46 |
+
],
|
| 47 |
+
"padding": "max_length",
|
| 48 |
+
"patch_size": 4,
|
| 49 |
+
"path_norm": true,
|
| 50 |
+
"problem_type": "single_label_classification",
|
| 51 |
+
"qkv_bias": true,
|
| 52 |
+
"stage_names": [
|
| 53 |
+
"stem",
|
| 54 |
+
"stage1",
|
| 55 |
+
"stage2",
|
| 56 |
+
"stage3",
|
| 57 |
+
"stage4"
|
| 58 |
+
],
|
| 59 |
+
"torch_dtype": "float32",
|
| 60 |
+
"transformers_version": "4.37.2",
|
| 61 |
+
"use_absolute_embeddings": false,
|
| 62 |
+
"window_size": 7
|
| 63 |
+
}
|
research/ai_detector/checkpoint-562/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9c54f1d503fc157004c496f277f4f6f5612c61527096bfb4b8afb7d5a18df36
|
| 3 |
+
size 347498816
|
research/ai_detector/checkpoint-562/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bb5c0e90ba3073b92823ea93c8ef1fe01f6e396117253d69b9cd60966177348
|
| 3 |
+
size 694317645
|
research/ai_detector/checkpoint-562/preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
research/ai_detector/checkpoint-562/rng_state.pth
ADDED
|
Binary file (14.2 kB). View file
|
|
|
research/ai_detector/checkpoint-562/scheduler.pt
ADDED
|
Binary file (1.06 kB). View file
|
|
|
research/ai_detector/checkpoint-562/trainer_state.json
ADDED
|
@@ -0,0 +1,375 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9964804552942939,
|
| 3 |
+
"best_model_checkpoint": "ai_detector/checkpoint-281",
|
| 4 |
+
"epoch": 1.9982222222222221,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 562,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"learning_rate": 5.882352941176471e-06,
|
| 14 |
+
"loss": 0.6876,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.07,
|
| 19 |
+
"learning_rate": 1.1764705882352942e-05,
|
| 20 |
+
"loss": 0.4009,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.11,
|
| 25 |
+
"learning_rate": 1.7647058823529414e-05,
|
| 26 |
+
"loss": 0.2111,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.14,
|
| 31 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 32 |
+
"loss": 0.0845,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.18,
|
| 37 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 38 |
+
"loss": 0.052,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.21,
|
| 43 |
+
"learning_rate": 3.529411764705883e-05,
|
| 44 |
+
"loss": 0.0831,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.25,
|
| 49 |
+
"learning_rate": 4.11764705882353e-05,
|
| 50 |
+
"loss": 0.0719,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.28,
|
| 55 |
+
"learning_rate": 4.705882352941177e-05,
|
| 56 |
+
"loss": 0.0493,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.32,
|
| 61 |
+
"learning_rate": 4.967018469656992e-05,
|
| 62 |
+
"loss": 0.0278,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.36,
|
| 67 |
+
"learning_rate": 4.901055408970976e-05,
|
| 68 |
+
"loss": 0.0447,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.39,
|
| 73 |
+
"learning_rate": 4.835092348284961e-05,
|
| 74 |
+
"loss": 0.0597,
|
| 75 |
+
"step": 110
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.43,
|
| 79 |
+
"learning_rate": 4.7691292875989446e-05,
|
| 80 |
+
"loss": 0.0194,
|
| 81 |
+
"step": 120
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.46,
|
| 85 |
+
"learning_rate": 4.703166226912929e-05,
|
| 86 |
+
"loss": 0.0308,
|
| 87 |
+
"step": 130
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.5,
|
| 91 |
+
"learning_rate": 4.6372031662269136e-05,
|
| 92 |
+
"loss": 0.0328,
|
| 93 |
+
"step": 140
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.53,
|
| 97 |
+
"learning_rate": 4.5712401055408974e-05,
|
| 98 |
+
"loss": 0.0415,
|
| 99 |
+
"step": 150
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.57,
|
| 103 |
+
"learning_rate": 4.505277044854881e-05,
|
| 104 |
+
"loss": 0.0075,
|
| 105 |
+
"step": 160
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.6,
|
| 109 |
+
"learning_rate": 4.439313984168866e-05,
|
| 110 |
+
"loss": 0.0358,
|
| 111 |
+
"step": 170
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.64,
|
| 115 |
+
"learning_rate": 4.3733509234828496e-05,
|
| 116 |
+
"loss": 0.0684,
|
| 117 |
+
"step": 180
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.68,
|
| 121 |
+
"learning_rate": 4.307387862796834e-05,
|
| 122 |
+
"loss": 0.0382,
|
| 123 |
+
"step": 190
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.71,
|
| 127 |
+
"learning_rate": 4.2414248021108186e-05,
|
| 128 |
+
"loss": 0.0252,
|
| 129 |
+
"step": 200
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.75,
|
| 133 |
+
"learning_rate": 4.1754617414248024e-05,
|
| 134 |
+
"loss": 0.0356,
|
| 135 |
+
"step": 210
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.78,
|
| 139 |
+
"learning_rate": 4.109498680738786e-05,
|
| 140 |
+
"loss": 0.0214,
|
| 141 |
+
"step": 220
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.82,
|
| 145 |
+
"learning_rate": 4.043535620052771e-05,
|
| 146 |
+
"loss": 0.0137,
|
| 147 |
+
"step": 230
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.85,
|
| 151 |
+
"learning_rate": 3.9775725593667545e-05,
|
| 152 |
+
"loss": 0.091,
|
| 153 |
+
"step": 240
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.89,
|
| 157 |
+
"learning_rate": 3.911609498680739e-05,
|
| 158 |
+
"loss": 0.0103,
|
| 159 |
+
"step": 250
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.92,
|
| 163 |
+
"learning_rate": 3.8456464379947235e-05,
|
| 164 |
+
"loss": 0.0418,
|
| 165 |
+
"step": 260
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.96,
|
| 169 |
+
"learning_rate": 3.779683377308707e-05,
|
| 170 |
+
"loss": 0.0089,
|
| 171 |
+
"step": 270
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.0,
|
| 175 |
+
"learning_rate": 3.713720316622691e-05,
|
| 176 |
+
"loss": 0.0347,
|
| 177 |
+
"step": 280
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.0,
|
| 181 |
+
"eval_accuracy": 0.9964804552942939,
|
| 182 |
+
"eval_loss": 0.013468941673636436,
|
| 183 |
+
"eval_runtime": 252.3562,
|
| 184 |
+
"eval_samples_per_second": 52.917,
|
| 185 |
+
"eval_steps_per_second": 3.309,
|
| 186 |
+
"step": 281
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.03,
|
| 190 |
+
"learning_rate": 3.6477572559366756e-05,
|
| 191 |
+
"loss": 0.0258,
|
| 192 |
+
"step": 290
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.07,
|
| 196 |
+
"learning_rate": 3.58179419525066e-05,
|
| 197 |
+
"loss": 0.0297,
|
| 198 |
+
"step": 300
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 1.1,
|
| 202 |
+
"learning_rate": 3.515831134564644e-05,
|
| 203 |
+
"loss": 0.029,
|
| 204 |
+
"step": 310
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.14,
|
| 208 |
+
"learning_rate": 3.4498680738786285e-05,
|
| 209 |
+
"loss": 0.0023,
|
| 210 |
+
"step": 320
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 1.17,
|
| 214 |
+
"learning_rate": 3.383905013192612e-05,
|
| 215 |
+
"loss": 0.013,
|
| 216 |
+
"step": 330
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.21,
|
| 220 |
+
"learning_rate": 3.317941952506596e-05,
|
| 221 |
+
"loss": 0.0203,
|
| 222 |
+
"step": 340
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.24,
|
| 226 |
+
"learning_rate": 3.2519788918205806e-05,
|
| 227 |
+
"loss": 0.0286,
|
| 228 |
+
"step": 350
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.28,
|
| 232 |
+
"learning_rate": 3.186015831134565e-05,
|
| 233 |
+
"loss": 0.0236,
|
| 234 |
+
"step": 360
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.32,
|
| 238 |
+
"learning_rate": 3.120052770448549e-05,
|
| 239 |
+
"loss": 0.0126,
|
| 240 |
+
"step": 370
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.35,
|
| 244 |
+
"learning_rate": 3.0540897097625334e-05,
|
| 245 |
+
"loss": 0.02,
|
| 246 |
+
"step": 380
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.39,
|
| 250 |
+
"learning_rate": 2.9881266490765176e-05,
|
| 251 |
+
"loss": 0.0215,
|
| 252 |
+
"step": 390
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.42,
|
| 256 |
+
"learning_rate": 2.9221635883905014e-05,
|
| 257 |
+
"loss": 0.0143,
|
| 258 |
+
"step": 400
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 1.46,
|
| 262 |
+
"learning_rate": 2.8562005277044855e-05,
|
| 263 |
+
"loss": 0.0054,
|
| 264 |
+
"step": 410
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.49,
|
| 268 |
+
"learning_rate": 2.79023746701847e-05,
|
| 269 |
+
"loss": 0.0072,
|
| 270 |
+
"step": 420
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.53,
|
| 274 |
+
"learning_rate": 2.724274406332454e-05,
|
| 275 |
+
"loss": 0.005,
|
| 276 |
+
"step": 430
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.56,
|
| 280 |
+
"learning_rate": 2.658311345646438e-05,
|
| 281 |
+
"loss": 0.0142,
|
| 282 |
+
"step": 440
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.6,
|
| 286 |
+
"learning_rate": 2.5923482849604225e-05,
|
| 287 |
+
"loss": 0.0155,
|
| 288 |
+
"step": 450
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.64,
|
| 292 |
+
"learning_rate": 2.5263852242744063e-05,
|
| 293 |
+
"loss": 0.0109,
|
| 294 |
+
"step": 460
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.67,
|
| 298 |
+
"learning_rate": 2.4604221635883905e-05,
|
| 299 |
+
"loss": 0.0073,
|
| 300 |
+
"step": 470
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.71,
|
| 304 |
+
"learning_rate": 2.3944591029023746e-05,
|
| 305 |
+
"loss": 0.0126,
|
| 306 |
+
"step": 480
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.74,
|
| 310 |
+
"learning_rate": 2.328496042216359e-05,
|
| 311 |
+
"loss": 0.0154,
|
| 312 |
+
"step": 490
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.78,
|
| 316 |
+
"learning_rate": 2.262532981530343e-05,
|
| 317 |
+
"loss": 0.0065,
|
| 318 |
+
"step": 500
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.81,
|
| 322 |
+
"learning_rate": 2.196569920844327e-05,
|
| 323 |
+
"loss": 0.0275,
|
| 324 |
+
"step": 510
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.85,
|
| 328 |
+
"learning_rate": 2.1306068601583116e-05,
|
| 329 |
+
"loss": 0.0296,
|
| 330 |
+
"step": 520
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.88,
|
| 334 |
+
"learning_rate": 2.0646437994722954e-05,
|
| 335 |
+
"loss": 0.0043,
|
| 336 |
+
"step": 530
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.92,
|
| 340 |
+
"learning_rate": 1.9986807387862796e-05,
|
| 341 |
+
"loss": 0.0012,
|
| 342 |
+
"step": 540
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.96,
|
| 346 |
+
"learning_rate": 1.932717678100264e-05,
|
| 347 |
+
"loss": 0.0044,
|
| 348 |
+
"step": 550
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.99,
|
| 352 |
+
"learning_rate": 1.866754617414248e-05,
|
| 353 |
+
"loss": 0.0009,
|
| 354 |
+
"step": 560
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.0,
|
| 358 |
+
"eval_accuracy": 0.9942339373970346,
|
| 359 |
+
"eval_loss": 0.02113017998635769,
|
| 360 |
+
"eval_runtime": 251.3615,
|
| 361 |
+
"eval_samples_per_second": 53.127,
|
| 362 |
+
"eval_steps_per_second": 3.322,
|
| 363 |
+
"step": 562
|
| 364 |
+
}
|
| 365 |
+
],
|
| 366 |
+
"logging_steps": 10,
|
| 367 |
+
"max_steps": 843,
|
| 368 |
+
"num_input_tokens_seen": 0,
|
| 369 |
+
"num_train_epochs": 3,
|
| 370 |
+
"save_steps": 500,
|
| 371 |
+
"total_flos": 2.820440002609152e+18,
|
| 372 |
+
"train_batch_size": 16,
|
| 373 |
+
"trial_name": null,
|
| 374 |
+
"trial_params": null
|
| 375 |
+
}
|
research/ai_detector/checkpoint-562/training_args.bin
ADDED
|
Binary file (4.66 kB). View file
|
|
|
research/ai_detector/checkpoint-843/config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "umm-maybe/AI-image-detector",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SwinForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"depths": [
|
| 8 |
+
2,
|
| 9 |
+
2,
|
| 10 |
+
18,
|
| 11 |
+
2
|
| 12 |
+
],
|
| 13 |
+
"drop_path_rate": 0.1,
|
| 14 |
+
"embed_dim": 128,
|
| 15 |
+
"encoder_stride": 32,
|
| 16 |
+
"hidden_act": "gelu",
|
| 17 |
+
"hidden_dropout_prob": 0.0,
|
| 18 |
+
"hidden_size": 1024,
|
| 19 |
+
"id2label": {
|
| 20 |
+
"0": "ai_gen",
|
| 21 |
+
"1": "human"
|
| 22 |
+
},
|
| 23 |
+
"image_size": 224,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"label2id": {
|
| 26 |
+
"ai_gen": "0",
|
| 27 |
+
"human": "1"
|
| 28 |
+
},
|
| 29 |
+
"layer_norm_eps": 1e-05,
|
| 30 |
+
"max_length": 128,
|
| 31 |
+
"mlp_ratio": 4.0,
|
| 32 |
+
"model_type": "swin",
|
| 33 |
+
"num_channels": 3,
|
| 34 |
+
"num_heads": [
|
| 35 |
+
4,
|
| 36 |
+
8,
|
| 37 |
+
16,
|
| 38 |
+
32
|
| 39 |
+
],
|
| 40 |
+
"num_layers": 4,
|
| 41 |
+
"out_features": [
|
| 42 |
+
"stage4"
|
| 43 |
+
],
|
| 44 |
+
"out_indices": [
|
| 45 |
+
4
|
| 46 |
+
],
|
| 47 |
+
"padding": "max_length",
|
| 48 |
+
"patch_size": 4,
|
| 49 |
+
"path_norm": true,
|
| 50 |
+
"problem_type": "single_label_classification",
|
| 51 |
+
"qkv_bias": true,
|
| 52 |
+
"stage_names": [
|
| 53 |
+
"stem",
|
| 54 |
+
"stage1",
|
| 55 |
+
"stage2",
|
| 56 |
+
"stage3",
|
| 57 |
+
"stage4"
|
| 58 |
+
],
|
| 59 |
+
"torch_dtype": "float32",
|
| 60 |
+
"transformers_version": "4.37.2",
|
| 61 |
+
"use_absolute_embeddings": false,
|
| 62 |
+
"window_size": 7
|
| 63 |
+
}
|
research/ai_detector/checkpoint-843/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4fd37a2fe0e66c4eb2afe850d5514b90ab5ab8ad6dd35cfc36b2088766aae32
|
| 3 |
+
size 347498816
|
research/ai_detector/checkpoint-843/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4c022c912a00d486187e84b7e495af6dafbd3d78eaca3dfcd6a78f6ace13eb50
|
| 3 |
+
size 694317645
|
research/ai_detector/checkpoint-843/preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
research/ai_detector/checkpoint-843/rng_state.pth
ADDED
|
Binary file (14.2 kB). View file
|
|
|
research/ai_detector/checkpoint-843/scheduler.pt
ADDED
|
Binary file (1.06 kB). View file
|
|
|
research/ai_detector/checkpoint-843/trainer_state.json
ADDED
|
@@ -0,0 +1,552 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9975288303130149,
|
| 3 |
+
"best_model_checkpoint": "ai_detector/checkpoint-843",
|
| 4 |
+
"epoch": 2.997333333333333,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 843,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"learning_rate": 5.882352941176471e-06,
|
| 14 |
+
"loss": 0.6876,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.07,
|
| 19 |
+
"learning_rate": 1.1764705882352942e-05,
|
| 20 |
+
"loss": 0.4009,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.11,
|
| 25 |
+
"learning_rate": 1.7647058823529414e-05,
|
| 26 |
+
"loss": 0.2111,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.14,
|
| 31 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 32 |
+
"loss": 0.0845,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.18,
|
| 37 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 38 |
+
"loss": 0.052,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.21,
|
| 43 |
+
"learning_rate": 3.529411764705883e-05,
|
| 44 |
+
"loss": 0.0831,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.25,
|
| 49 |
+
"learning_rate": 4.11764705882353e-05,
|
| 50 |
+
"loss": 0.0719,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.28,
|
| 55 |
+
"learning_rate": 4.705882352941177e-05,
|
| 56 |
+
"loss": 0.0493,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.32,
|
| 61 |
+
"learning_rate": 4.967018469656992e-05,
|
| 62 |
+
"loss": 0.0278,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.36,
|
| 67 |
+
"learning_rate": 4.901055408970976e-05,
|
| 68 |
+
"loss": 0.0447,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.39,
|
| 73 |
+
"learning_rate": 4.835092348284961e-05,
|
| 74 |
+
"loss": 0.0597,
|
| 75 |
+
"step": 110
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.43,
|
| 79 |
+
"learning_rate": 4.7691292875989446e-05,
|
| 80 |
+
"loss": 0.0194,
|
| 81 |
+
"step": 120
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.46,
|
| 85 |
+
"learning_rate": 4.703166226912929e-05,
|
| 86 |
+
"loss": 0.0308,
|
| 87 |
+
"step": 130
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.5,
|
| 91 |
+
"learning_rate": 4.6372031662269136e-05,
|
| 92 |
+
"loss": 0.0328,
|
| 93 |
+
"step": 140
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.53,
|
| 97 |
+
"learning_rate": 4.5712401055408974e-05,
|
| 98 |
+
"loss": 0.0415,
|
| 99 |
+
"step": 150
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.57,
|
| 103 |
+
"learning_rate": 4.505277044854881e-05,
|
| 104 |
+
"loss": 0.0075,
|
| 105 |
+
"step": 160
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.6,
|
| 109 |
+
"learning_rate": 4.439313984168866e-05,
|
| 110 |
+
"loss": 0.0358,
|
| 111 |
+
"step": 170
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.64,
|
| 115 |
+
"learning_rate": 4.3733509234828496e-05,
|
| 116 |
+
"loss": 0.0684,
|
| 117 |
+
"step": 180
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.68,
|
| 121 |
+
"learning_rate": 4.307387862796834e-05,
|
| 122 |
+
"loss": 0.0382,
|
| 123 |
+
"step": 190
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.71,
|
| 127 |
+
"learning_rate": 4.2414248021108186e-05,
|
| 128 |
+
"loss": 0.0252,
|
| 129 |
+
"step": 200
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.75,
|
| 133 |
+
"learning_rate": 4.1754617414248024e-05,
|
| 134 |
+
"loss": 0.0356,
|
| 135 |
+
"step": 210
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.78,
|
| 139 |
+
"learning_rate": 4.109498680738786e-05,
|
| 140 |
+
"loss": 0.0214,
|
| 141 |
+
"step": 220
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.82,
|
| 145 |
+
"learning_rate": 4.043535620052771e-05,
|
| 146 |
+
"loss": 0.0137,
|
| 147 |
+
"step": 230
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.85,
|
| 151 |
+
"learning_rate": 3.9775725593667545e-05,
|
| 152 |
+
"loss": 0.091,
|
| 153 |
+
"step": 240
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.89,
|
| 157 |
+
"learning_rate": 3.911609498680739e-05,
|
| 158 |
+
"loss": 0.0103,
|
| 159 |
+
"step": 250
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.92,
|
| 163 |
+
"learning_rate": 3.8456464379947235e-05,
|
| 164 |
+
"loss": 0.0418,
|
| 165 |
+
"step": 260
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.96,
|
| 169 |
+
"learning_rate": 3.779683377308707e-05,
|
| 170 |
+
"loss": 0.0089,
|
| 171 |
+
"step": 270
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.0,
|
| 175 |
+
"learning_rate": 3.713720316622691e-05,
|
| 176 |
+
"loss": 0.0347,
|
| 177 |
+
"step": 280
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.0,
|
| 181 |
+
"eval_accuracy": 0.9964804552942939,
|
| 182 |
+
"eval_loss": 0.013468941673636436,
|
| 183 |
+
"eval_runtime": 252.3562,
|
| 184 |
+
"eval_samples_per_second": 52.917,
|
| 185 |
+
"eval_steps_per_second": 3.309,
|
| 186 |
+
"step": 281
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.03,
|
| 190 |
+
"learning_rate": 3.6477572559366756e-05,
|
| 191 |
+
"loss": 0.0258,
|
| 192 |
+
"step": 290
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.07,
|
| 196 |
+
"learning_rate": 3.58179419525066e-05,
|
| 197 |
+
"loss": 0.0297,
|
| 198 |
+
"step": 300
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 1.1,
|
| 202 |
+
"learning_rate": 3.515831134564644e-05,
|
| 203 |
+
"loss": 0.029,
|
| 204 |
+
"step": 310
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.14,
|
| 208 |
+
"learning_rate": 3.4498680738786285e-05,
|
| 209 |
+
"loss": 0.0023,
|
| 210 |
+
"step": 320
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 1.17,
|
| 214 |
+
"learning_rate": 3.383905013192612e-05,
|
| 215 |
+
"loss": 0.013,
|
| 216 |
+
"step": 330
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.21,
|
| 220 |
+
"learning_rate": 3.317941952506596e-05,
|
| 221 |
+
"loss": 0.0203,
|
| 222 |
+
"step": 340
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.24,
|
| 226 |
+
"learning_rate": 3.2519788918205806e-05,
|
| 227 |
+
"loss": 0.0286,
|
| 228 |
+
"step": 350
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.28,
|
| 232 |
+
"learning_rate": 3.186015831134565e-05,
|
| 233 |
+
"loss": 0.0236,
|
| 234 |
+
"step": 360
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.32,
|
| 238 |
+
"learning_rate": 3.120052770448549e-05,
|
| 239 |
+
"loss": 0.0126,
|
| 240 |
+
"step": 370
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.35,
|
| 244 |
+
"learning_rate": 3.0540897097625334e-05,
|
| 245 |
+
"loss": 0.02,
|
| 246 |
+
"step": 380
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.39,
|
| 250 |
+
"learning_rate": 2.9881266490765176e-05,
|
| 251 |
+
"loss": 0.0215,
|
| 252 |
+
"step": 390
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.42,
|
| 256 |
+
"learning_rate": 2.9221635883905014e-05,
|
| 257 |
+
"loss": 0.0143,
|
| 258 |
+
"step": 400
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 1.46,
|
| 262 |
+
"learning_rate": 2.8562005277044855e-05,
|
| 263 |
+
"loss": 0.0054,
|
| 264 |
+
"step": 410
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.49,
|
| 268 |
+
"learning_rate": 2.79023746701847e-05,
|
| 269 |
+
"loss": 0.0072,
|
| 270 |
+
"step": 420
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.53,
|
| 274 |
+
"learning_rate": 2.724274406332454e-05,
|
| 275 |
+
"loss": 0.005,
|
| 276 |
+
"step": 430
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.56,
|
| 280 |
+
"learning_rate": 2.658311345646438e-05,
|
| 281 |
+
"loss": 0.0142,
|
| 282 |
+
"step": 440
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.6,
|
| 286 |
+
"learning_rate": 2.5923482849604225e-05,
|
| 287 |
+
"loss": 0.0155,
|
| 288 |
+
"step": 450
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.64,
|
| 292 |
+
"learning_rate": 2.5263852242744063e-05,
|
| 293 |
+
"loss": 0.0109,
|
| 294 |
+
"step": 460
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.67,
|
| 298 |
+
"learning_rate": 2.4604221635883905e-05,
|
| 299 |
+
"loss": 0.0073,
|
| 300 |
+
"step": 470
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.71,
|
| 304 |
+
"learning_rate": 2.3944591029023746e-05,
|
| 305 |
+
"loss": 0.0126,
|
| 306 |
+
"step": 480
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.74,
|
| 310 |
+
"learning_rate": 2.328496042216359e-05,
|
| 311 |
+
"loss": 0.0154,
|
| 312 |
+
"step": 490
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.78,
|
| 316 |
+
"learning_rate": 2.262532981530343e-05,
|
| 317 |
+
"loss": 0.0065,
|
| 318 |
+
"step": 500
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.81,
|
| 322 |
+
"learning_rate": 2.196569920844327e-05,
|
| 323 |
+
"loss": 0.0275,
|
| 324 |
+
"step": 510
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.85,
|
| 328 |
+
"learning_rate": 2.1306068601583116e-05,
|
| 329 |
+
"loss": 0.0296,
|
| 330 |
+
"step": 520
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.88,
|
| 334 |
+
"learning_rate": 2.0646437994722954e-05,
|
| 335 |
+
"loss": 0.0043,
|
| 336 |
+
"step": 530
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.92,
|
| 340 |
+
"learning_rate": 1.9986807387862796e-05,
|
| 341 |
+
"loss": 0.0012,
|
| 342 |
+
"step": 540
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.96,
|
| 346 |
+
"learning_rate": 1.932717678100264e-05,
|
| 347 |
+
"loss": 0.0044,
|
| 348 |
+
"step": 550
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.99,
|
| 352 |
+
"learning_rate": 1.866754617414248e-05,
|
| 353 |
+
"loss": 0.0009,
|
| 354 |
+
"step": 560
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.0,
|
| 358 |
+
"eval_accuracy": 0.9942339373970346,
|
| 359 |
+
"eval_loss": 0.02113017998635769,
|
| 360 |
+
"eval_runtime": 251.3615,
|
| 361 |
+
"eval_samples_per_second": 53.127,
|
| 362 |
+
"eval_steps_per_second": 3.322,
|
| 363 |
+
"step": 562
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.03,
|
| 367 |
+
"learning_rate": 1.8007915567282324e-05,
|
| 368 |
+
"loss": 0.002,
|
| 369 |
+
"step": 570
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.06,
|
| 373 |
+
"learning_rate": 1.7348284960422166e-05,
|
| 374 |
+
"loss": 0.0154,
|
| 375 |
+
"step": 580
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.1,
|
| 379 |
+
"learning_rate": 1.6688654353562007e-05,
|
| 380 |
+
"loss": 0.0058,
|
| 381 |
+
"step": 590
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 2.13,
|
| 385 |
+
"learning_rate": 1.602902374670185e-05,
|
| 386 |
+
"loss": 0.0066,
|
| 387 |
+
"step": 600
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 2.17,
|
| 391 |
+
"learning_rate": 1.536939313984169e-05,
|
| 392 |
+
"loss": 0.0007,
|
| 393 |
+
"step": 610
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 2.2,
|
| 397 |
+
"learning_rate": 1.470976253298153e-05,
|
| 398 |
+
"loss": 0.0008,
|
| 399 |
+
"step": 620
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.24,
|
| 403 |
+
"learning_rate": 1.4050131926121373e-05,
|
| 404 |
+
"loss": 0.0037,
|
| 405 |
+
"step": 630
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 2.28,
|
| 409 |
+
"learning_rate": 1.3390501319261215e-05,
|
| 410 |
+
"loss": 0.0057,
|
| 411 |
+
"step": 640
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 2.31,
|
| 415 |
+
"learning_rate": 1.2730870712401055e-05,
|
| 416 |
+
"loss": 0.0002,
|
| 417 |
+
"step": 650
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.35,
|
| 421 |
+
"learning_rate": 1.2071240105540896e-05,
|
| 422 |
+
"loss": 0.0241,
|
| 423 |
+
"step": 660
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 2.38,
|
| 427 |
+
"learning_rate": 1.141160949868074e-05,
|
| 428 |
+
"loss": 0.0007,
|
| 429 |
+
"step": 670
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.42,
|
| 433 |
+
"learning_rate": 1.0751978891820581e-05,
|
| 434 |
+
"loss": 0.0058,
|
| 435 |
+
"step": 680
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 2.45,
|
| 439 |
+
"learning_rate": 1.0092348284960421e-05,
|
| 440 |
+
"loss": 0.0003,
|
| 441 |
+
"step": 690
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 2.49,
|
| 445 |
+
"learning_rate": 9.432717678100264e-06,
|
| 446 |
+
"loss": 0.0075,
|
| 447 |
+
"step": 700
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 2.52,
|
| 451 |
+
"learning_rate": 8.773087071240106e-06,
|
| 452 |
+
"loss": 0.001,
|
| 453 |
+
"step": 710
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 2.56,
|
| 457 |
+
"learning_rate": 8.113456464379948e-06,
|
| 458 |
+
"loss": 0.002,
|
| 459 |
+
"step": 720
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.6,
|
| 463 |
+
"learning_rate": 7.453825857519789e-06,
|
| 464 |
+
"loss": 0.0125,
|
| 465 |
+
"step": 730
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 2.63,
|
| 469 |
+
"learning_rate": 6.794195250659631e-06,
|
| 470 |
+
"loss": 0.0001,
|
| 471 |
+
"step": 740
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 2.67,
|
| 475 |
+
"learning_rate": 6.134564643799472e-06,
|
| 476 |
+
"loss": 0.0003,
|
| 477 |
+
"step": 750
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 2.7,
|
| 481 |
+
"learning_rate": 5.474934036939315e-06,
|
| 482 |
+
"loss": 0.001,
|
| 483 |
+
"step": 760
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 2.74,
|
| 487 |
+
"learning_rate": 4.8153034300791555e-06,
|
| 488 |
+
"loss": 0.0001,
|
| 489 |
+
"step": 770
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.77,
|
| 493 |
+
"learning_rate": 4.155672823218998e-06,
|
| 494 |
+
"loss": 0.0078,
|
| 495 |
+
"step": 780
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 2.81,
|
| 499 |
+
"learning_rate": 3.496042216358839e-06,
|
| 500 |
+
"loss": 0.0031,
|
| 501 |
+
"step": 790
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.84,
|
| 505 |
+
"learning_rate": 2.836411609498681e-06,
|
| 506 |
+
"loss": 0.0264,
|
| 507 |
+
"step": 800
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 2.88,
|
| 511 |
+
"learning_rate": 2.1767810026385226e-06,
|
| 512 |
+
"loss": 0.0124,
|
| 513 |
+
"step": 810
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 2.92,
|
| 517 |
+
"learning_rate": 1.5171503957783642e-06,
|
| 518 |
+
"loss": 0.0195,
|
| 519 |
+
"step": 820
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 2.95,
|
| 523 |
+
"learning_rate": 8.575197889182058e-07,
|
| 524 |
+
"loss": 0.005,
|
| 525 |
+
"step": 830
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 2.99,
|
| 529 |
+
"learning_rate": 1.9788918205804752e-07,
|
| 530 |
+
"loss": 0.0241,
|
| 531 |
+
"step": 840
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 3.0,
|
| 535 |
+
"eval_accuracy": 0.9975288303130149,
|
| 536 |
+
"eval_loss": 0.010735484771430492,
|
| 537 |
+
"eval_runtime": 284.9371,
|
| 538 |
+
"eval_samples_per_second": 46.866,
|
| 539 |
+
"eval_steps_per_second": 2.93,
|
| 540 |
+
"step": 843
|
| 541 |
+
}
|
| 542 |
+
],
|
| 543 |
+
"logging_steps": 10,
|
| 544 |
+
"max_steps": 843,
|
| 545 |
+
"num_input_tokens_seen": 0,
|
| 546 |
+
"num_train_epochs": 3,
|
| 547 |
+
"save_steps": 500,
|
| 548 |
+
"total_flos": 4.2268994172435825e+18,
|
| 549 |
+
"train_batch_size": 16,
|
| 550 |
+
"trial_name": null,
|
| 551 |
+
"trial_params": null
|
| 552 |
+
}
|
research/ai_detector/checkpoint-843/training_args.bin
ADDED
|
Binary file (4.66 kB). View file
|
|
|
research/ai_detector_v2/checkpoint-1210/config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "umm-maybe/AI-image-detector",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SwinForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"depths": [
|
| 8 |
+
2,
|
| 9 |
+
2,
|
| 10 |
+
18,
|
| 11 |
+
2
|
| 12 |
+
],
|
| 13 |
+
"drop_path_rate": 0.1,
|
| 14 |
+
"embed_dim": 128,
|
| 15 |
+
"encoder_stride": 32,
|
| 16 |
+
"hidden_act": "gelu",
|
| 17 |
+
"hidden_dropout_prob": 0.0,
|
| 18 |
+
"hidden_size": 1024,
|
| 19 |
+
"id2label": {
|
| 20 |
+
"0": "ai_gen",
|
| 21 |
+
"1": "human"
|
| 22 |
+
},
|
| 23 |
+
"image_size": 224,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"label2id": {
|
| 26 |
+
"ai_gen": "0",
|
| 27 |
+
"human": "1"
|
| 28 |
+
},
|
| 29 |
+
"layer_norm_eps": 1e-05,
|
| 30 |
+
"max_length": 128,
|
| 31 |
+
"mlp_ratio": 4.0,
|
| 32 |
+
"model_type": "swin",
|
| 33 |
+
"num_channels": 3,
|
| 34 |
+
"num_heads": [
|
| 35 |
+
4,
|
| 36 |
+
8,
|
| 37 |
+
16,
|
| 38 |
+
32
|
| 39 |
+
],
|
| 40 |
+
"num_layers": 4,
|
| 41 |
+
"out_features": [
|
| 42 |
+
"stage4"
|
| 43 |
+
],
|
| 44 |
+
"out_indices": [
|
| 45 |
+
4
|
| 46 |
+
],
|
| 47 |
+
"padding": "max_length",
|
| 48 |
+
"patch_size": 4,
|
| 49 |
+
"path_norm": true,
|
| 50 |
+
"problem_type": "single_label_classification",
|
| 51 |
+
"qkv_bias": true,
|
| 52 |
+
"stage_names": [
|
| 53 |
+
"stem",
|
| 54 |
+
"stage1",
|
| 55 |
+
"stage2",
|
| 56 |
+
"stage3",
|
| 57 |
+
"stage4"
|
| 58 |
+
],
|
| 59 |
+
"torch_dtype": "float32",
|
| 60 |
+
"transformers_version": "4.37.2",
|
| 61 |
+
"use_absolute_embeddings": false,
|
| 62 |
+
"window_size": 7
|
| 63 |
+
}
|
research/ai_detector_v2/checkpoint-1210/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:56c806ccd482578edc7d4bdddd66e768243a958ccef4dacca5a40f90a996175c
|
| 3 |
+
size 347498816
|
research/ai_detector_v2/checkpoint-1210/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8a354c6fadbd0a0cd8312faca439578b35e360eb92978f86758071123c0a0cb9
|
| 3 |
+
size 694317645
|
research/ai_detector_v2/checkpoint-1210/preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
research/ai_detector_v2/checkpoint-1210/rng_state.pth
ADDED
|
Binary file (14.2 kB). View file
|
|
|
research/ai_detector_v2/checkpoint-1210/scheduler.pt
ADDED
|
Binary file (1.06 kB). View file
|
|
|
research/ai_detector_v2/checkpoint-1210/trainer_state.json
ADDED
|
@@ -0,0 +1,765 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9980530178223753,
|
| 3 |
+
"best_model_checkpoint": "ai_detector_v2/checkpoint-1210",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1210,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 2.747252747252747e-06,
|
| 14 |
+
"loss": 0.7391,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.03,
|
| 19 |
+
"learning_rate": 5.494505494505494e-06,
|
| 20 |
+
"loss": 0.4175,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.05,
|
| 25 |
+
"learning_rate": 8.241758241758243e-06,
|
| 26 |
+
"loss": 0.3476,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.07,
|
| 31 |
+
"learning_rate": 1.0989010989010989e-05,
|
| 32 |
+
"loss": 0.1961,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.08,
|
| 37 |
+
"learning_rate": 1.3736263736263738e-05,
|
| 38 |
+
"loss": 0.1436,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.1,
|
| 43 |
+
"learning_rate": 1.6483516483516486e-05,
|
| 44 |
+
"loss": 0.0595,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.12,
|
| 49 |
+
"learning_rate": 1.923076923076923e-05,
|
| 50 |
+
"loss": 0.0585,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.13,
|
| 55 |
+
"learning_rate": 2.1978021978021977e-05,
|
| 56 |
+
"loss": 0.0607,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.15,
|
| 61 |
+
"learning_rate": 2.4725274725274727e-05,
|
| 62 |
+
"loss": 0.041,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.17,
|
| 67 |
+
"learning_rate": 2.7472527472527476e-05,
|
| 68 |
+
"loss": 0.0561,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.18,
|
| 73 |
+
"learning_rate": 3.021978021978022e-05,
|
| 74 |
+
"loss": 0.0517,
|
| 75 |
+
"step": 110
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.2,
|
| 79 |
+
"learning_rate": 3.296703296703297e-05,
|
| 80 |
+
"loss": 0.0581,
|
| 81 |
+
"step": 120
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.21,
|
| 85 |
+
"learning_rate": 3.571428571428572e-05,
|
| 86 |
+
"loss": 0.0281,
|
| 87 |
+
"step": 130
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.23,
|
| 91 |
+
"learning_rate": 3.846153846153846e-05,
|
| 92 |
+
"loss": 0.0612,
|
| 93 |
+
"step": 140
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.25,
|
| 97 |
+
"learning_rate": 4.120879120879121e-05,
|
| 98 |
+
"loss": 0.0297,
|
| 99 |
+
"step": 150
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.26,
|
| 103 |
+
"learning_rate": 4.3956043956043955e-05,
|
| 104 |
+
"loss": 0.021,
|
| 105 |
+
"step": 160
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.28,
|
| 109 |
+
"learning_rate": 4.670329670329671e-05,
|
| 110 |
+
"loss": 0.019,
|
| 111 |
+
"step": 170
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.3,
|
| 115 |
+
"learning_rate": 4.945054945054945e-05,
|
| 116 |
+
"loss": 0.0233,
|
| 117 |
+
"step": 180
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.31,
|
| 121 |
+
"learning_rate": 4.9755052051439074e-05,
|
| 122 |
+
"loss": 0.0298,
|
| 123 |
+
"step": 190
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.33,
|
| 127 |
+
"learning_rate": 4.944886711573791e-05,
|
| 128 |
+
"loss": 0.0421,
|
| 129 |
+
"step": 200
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.35,
|
| 133 |
+
"learning_rate": 4.914268218003674e-05,
|
| 134 |
+
"loss": 0.0277,
|
| 135 |
+
"step": 210
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.36,
|
| 139 |
+
"learning_rate": 4.883649724433558e-05,
|
| 140 |
+
"loss": 0.0361,
|
| 141 |
+
"step": 220
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.38,
|
| 145 |
+
"learning_rate": 4.8530312308634416e-05,
|
| 146 |
+
"loss": 0.0579,
|
| 147 |
+
"step": 230
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.4,
|
| 151 |
+
"learning_rate": 4.822412737293325e-05,
|
| 152 |
+
"loss": 0.0338,
|
| 153 |
+
"step": 240
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.41,
|
| 157 |
+
"learning_rate": 4.791794243723209e-05,
|
| 158 |
+
"loss": 0.0366,
|
| 159 |
+
"step": 250
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.43,
|
| 163 |
+
"learning_rate": 4.761175750153093e-05,
|
| 164 |
+
"loss": 0.0092,
|
| 165 |
+
"step": 260
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.45,
|
| 169 |
+
"learning_rate": 4.7305572565829766e-05,
|
| 170 |
+
"loss": 0.063,
|
| 171 |
+
"step": 270
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.46,
|
| 175 |
+
"learning_rate": 4.69993876301286e-05,
|
| 176 |
+
"loss": 0.0527,
|
| 177 |
+
"step": 280
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.48,
|
| 181 |
+
"learning_rate": 4.6693202694427433e-05,
|
| 182 |
+
"loss": 0.023,
|
| 183 |
+
"step": 290
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.5,
|
| 187 |
+
"learning_rate": 4.6387017758726274e-05,
|
| 188 |
+
"loss": 0.0437,
|
| 189 |
+
"step": 300
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.51,
|
| 193 |
+
"learning_rate": 4.608083282302511e-05,
|
| 194 |
+
"loss": 0.0381,
|
| 195 |
+
"step": 310
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.53,
|
| 199 |
+
"learning_rate": 4.577464788732395e-05,
|
| 200 |
+
"loss": 0.0288,
|
| 201 |
+
"step": 320
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.55,
|
| 205 |
+
"learning_rate": 4.546846295162278e-05,
|
| 206 |
+
"loss": 0.0385,
|
| 207 |
+
"step": 330
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.56,
|
| 211 |
+
"learning_rate": 4.516227801592162e-05,
|
| 212 |
+
"loss": 0.0442,
|
| 213 |
+
"step": 340
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.58,
|
| 217 |
+
"learning_rate": 4.485609308022046e-05,
|
| 218 |
+
"loss": 0.0128,
|
| 219 |
+
"step": 350
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6,
|
| 223 |
+
"learning_rate": 4.454990814451929e-05,
|
| 224 |
+
"loss": 0.0189,
|
| 225 |
+
"step": 360
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.61,
|
| 229 |
+
"learning_rate": 4.4243723208818125e-05,
|
| 230 |
+
"loss": 0.0219,
|
| 231 |
+
"step": 370
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.63,
|
| 235 |
+
"learning_rate": 4.3937538273116966e-05,
|
| 236 |
+
"loss": 0.0296,
|
| 237 |
+
"step": 380
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.64,
|
| 241 |
+
"learning_rate": 4.363135333741581e-05,
|
| 242 |
+
"loss": 0.058,
|
| 243 |
+
"step": 390
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.66,
|
| 247 |
+
"learning_rate": 4.332516840171464e-05,
|
| 248 |
+
"loss": 0.0317,
|
| 249 |
+
"step": 400
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.68,
|
| 253 |
+
"learning_rate": 4.3018983466013475e-05,
|
| 254 |
+
"loss": 0.0364,
|
| 255 |
+
"step": 410
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.69,
|
| 259 |
+
"learning_rate": 4.271279853031231e-05,
|
| 260 |
+
"loss": 0.0239,
|
| 261 |
+
"step": 420
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.71,
|
| 265 |
+
"learning_rate": 4.240661359461115e-05,
|
| 266 |
+
"loss": 0.0265,
|
| 267 |
+
"step": 430
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.73,
|
| 271 |
+
"learning_rate": 4.210042865890998e-05,
|
| 272 |
+
"loss": 0.0046,
|
| 273 |
+
"step": 440
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.74,
|
| 277 |
+
"learning_rate": 4.179424372320882e-05,
|
| 278 |
+
"loss": 0.0182,
|
| 279 |
+
"step": 450
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.76,
|
| 283 |
+
"learning_rate": 4.148805878750766e-05,
|
| 284 |
+
"loss": 0.0255,
|
| 285 |
+
"step": 460
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.78,
|
| 289 |
+
"learning_rate": 4.118187385180649e-05,
|
| 290 |
+
"loss": 0.0143,
|
| 291 |
+
"step": 470
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.79,
|
| 295 |
+
"learning_rate": 4.087568891610533e-05,
|
| 296 |
+
"loss": 0.0348,
|
| 297 |
+
"step": 480
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.81,
|
| 301 |
+
"learning_rate": 4.0569503980404167e-05,
|
| 302 |
+
"loss": 0.035,
|
| 303 |
+
"step": 490
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.83,
|
| 307 |
+
"learning_rate": 4.0263319044703e-05,
|
| 308 |
+
"loss": 0.0397,
|
| 309 |
+
"step": 500
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.84,
|
| 313 |
+
"learning_rate": 3.9957134109001834e-05,
|
| 314 |
+
"loss": 0.0194,
|
| 315 |
+
"step": 510
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.86,
|
| 319 |
+
"learning_rate": 3.9650949173300675e-05,
|
| 320 |
+
"loss": 0.0075,
|
| 321 |
+
"step": 520
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.88,
|
| 325 |
+
"learning_rate": 3.9344764237599516e-05,
|
| 326 |
+
"loss": 0.0025,
|
| 327 |
+
"step": 530
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.89,
|
| 331 |
+
"learning_rate": 3.903857930189835e-05,
|
| 332 |
+
"loss": 0.002,
|
| 333 |
+
"step": 540
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.91,
|
| 337 |
+
"learning_rate": 3.8732394366197184e-05,
|
| 338 |
+
"loss": 0.0093,
|
| 339 |
+
"step": 550
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.93,
|
| 343 |
+
"learning_rate": 3.8426209430496025e-05,
|
| 344 |
+
"loss": 0.0027,
|
| 345 |
+
"step": 560
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.94,
|
| 349 |
+
"learning_rate": 3.812002449479486e-05,
|
| 350 |
+
"loss": 0.011,
|
| 351 |
+
"step": 570
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 0.96,
|
| 355 |
+
"learning_rate": 3.781383955909369e-05,
|
| 356 |
+
"loss": 0.0133,
|
| 357 |
+
"step": 580
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.98,
|
| 361 |
+
"learning_rate": 3.7507654623392526e-05,
|
| 362 |
+
"loss": 0.0314,
|
| 363 |
+
"step": 590
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.99,
|
| 367 |
+
"learning_rate": 3.720146968769137e-05,
|
| 368 |
+
"loss": 0.0187,
|
| 369 |
+
"step": 600
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.0,
|
| 373 |
+
"eval_accuracy": 0.9973790624531975,
|
| 374 |
+
"eval_loss": 0.010242484509944916,
|
| 375 |
+
"eval_runtime": 251.9013,
|
| 376 |
+
"eval_samples_per_second": 53.013,
|
| 377 |
+
"eval_steps_per_second": 3.315,
|
| 378 |
+
"step": 605
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.01,
|
| 382 |
+
"learning_rate": 3.689528475199021e-05,
|
| 383 |
+
"loss": 0.0335,
|
| 384 |
+
"step": 610
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.02,
|
| 388 |
+
"learning_rate": 3.658909981628904e-05,
|
| 389 |
+
"loss": 0.0224,
|
| 390 |
+
"step": 620
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.04,
|
| 394 |
+
"learning_rate": 3.6282914880587876e-05,
|
| 395 |
+
"loss": 0.0192,
|
| 396 |
+
"step": 630
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.06,
|
| 400 |
+
"learning_rate": 3.597672994488671e-05,
|
| 401 |
+
"loss": 0.0164,
|
| 402 |
+
"step": 640
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.07,
|
| 406 |
+
"learning_rate": 3.567054500918555e-05,
|
| 407 |
+
"loss": 0.0109,
|
| 408 |
+
"step": 650
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.09,
|
| 412 |
+
"learning_rate": 3.5364360073484384e-05,
|
| 413 |
+
"loss": 0.013,
|
| 414 |
+
"step": 660
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.11,
|
| 418 |
+
"learning_rate": 3.5058175137783225e-05,
|
| 419 |
+
"loss": 0.0055,
|
| 420 |
+
"step": 670
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 1.12,
|
| 424 |
+
"learning_rate": 3.475199020208206e-05,
|
| 425 |
+
"loss": 0.0191,
|
| 426 |
+
"step": 680
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 1.14,
|
| 430 |
+
"learning_rate": 3.44458052663809e-05,
|
| 431 |
+
"loss": 0.0013,
|
| 432 |
+
"step": 690
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 1.16,
|
| 436 |
+
"learning_rate": 3.4139620330679734e-05,
|
| 437 |
+
"loss": 0.0186,
|
| 438 |
+
"step": 700
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 1.17,
|
| 442 |
+
"learning_rate": 3.383343539497857e-05,
|
| 443 |
+
"loss": 0.0001,
|
| 444 |
+
"step": 710
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.19,
|
| 448 |
+
"learning_rate": 3.35272504592774e-05,
|
| 449 |
+
"loss": 0.0127,
|
| 450 |
+
"step": 720
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.21,
|
| 454 |
+
"learning_rate": 3.322106552357624e-05,
|
| 455 |
+
"loss": 0.0052,
|
| 456 |
+
"step": 730
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 1.22,
|
| 460 |
+
"learning_rate": 3.291488058787508e-05,
|
| 461 |
+
"loss": 0.0278,
|
| 462 |
+
"step": 740
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 1.24,
|
| 466 |
+
"learning_rate": 3.260869565217392e-05,
|
| 467 |
+
"loss": 0.0015,
|
| 468 |
+
"step": 750
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 1.26,
|
| 472 |
+
"learning_rate": 3.230251071647275e-05,
|
| 473 |
+
"loss": 0.0026,
|
| 474 |
+
"step": 760
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.27,
|
| 478 |
+
"learning_rate": 3.1996325780771585e-05,
|
| 479 |
+
"loss": 0.0243,
|
| 480 |
+
"step": 770
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 1.29,
|
| 484 |
+
"learning_rate": 3.1690140845070426e-05,
|
| 485 |
+
"loss": 0.0142,
|
| 486 |
+
"step": 780
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.31,
|
| 490 |
+
"learning_rate": 3.138395590936926e-05,
|
| 491 |
+
"loss": 0.007,
|
| 492 |
+
"step": 790
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.32,
|
| 496 |
+
"learning_rate": 3.1077770973668093e-05,
|
| 497 |
+
"loss": 0.0175,
|
| 498 |
+
"step": 800
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 1.34,
|
| 502 |
+
"learning_rate": 3.0771586037966934e-05,
|
| 503 |
+
"loss": 0.0064,
|
| 504 |
+
"step": 810
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.36,
|
| 508 |
+
"learning_rate": 3.046540110226577e-05,
|
| 509 |
+
"loss": 0.0168,
|
| 510 |
+
"step": 820
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 1.37,
|
| 514 |
+
"learning_rate": 3.015921616656461e-05,
|
| 515 |
+
"loss": 0.002,
|
| 516 |
+
"step": 830
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 1.39,
|
| 520 |
+
"learning_rate": 2.9853031230863443e-05,
|
| 521 |
+
"loss": 0.0129,
|
| 522 |
+
"step": 840
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 1.4,
|
| 526 |
+
"learning_rate": 2.9546846295162277e-05,
|
| 527 |
+
"loss": 0.0134,
|
| 528 |
+
"step": 850
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.42,
|
| 532 |
+
"learning_rate": 2.9240661359461114e-05,
|
| 533 |
+
"loss": 0.0194,
|
| 534 |
+
"step": 860
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.44,
|
| 538 |
+
"learning_rate": 2.8934476423759955e-05,
|
| 539 |
+
"loss": 0.0096,
|
| 540 |
+
"step": 870
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 1.45,
|
| 544 |
+
"learning_rate": 2.862829148805879e-05,
|
| 545 |
+
"loss": 0.0048,
|
| 546 |
+
"step": 880
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.47,
|
| 550 |
+
"learning_rate": 2.8322106552357626e-05,
|
| 551 |
+
"loss": 0.0003,
|
| 552 |
+
"step": 890
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 1.49,
|
| 556 |
+
"learning_rate": 2.801592161665646e-05,
|
| 557 |
+
"loss": 0.0115,
|
| 558 |
+
"step": 900
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 1.5,
|
| 562 |
+
"learning_rate": 2.77097366809553e-05,
|
| 563 |
+
"loss": 0.0264,
|
| 564 |
+
"step": 910
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.52,
|
| 568 |
+
"learning_rate": 2.7403551745254135e-05,
|
| 569 |
+
"loss": 0.0233,
|
| 570 |
+
"step": 920
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.54,
|
| 574 |
+
"learning_rate": 2.7097366809552972e-05,
|
| 575 |
+
"loss": 0.0008,
|
| 576 |
+
"step": 930
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.55,
|
| 580 |
+
"learning_rate": 2.6791181873851806e-05,
|
| 581 |
+
"loss": 0.0103,
|
| 582 |
+
"step": 940
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 1.57,
|
| 586 |
+
"learning_rate": 2.6484996938150647e-05,
|
| 587 |
+
"loss": 0.0356,
|
| 588 |
+
"step": 950
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 1.59,
|
| 592 |
+
"learning_rate": 2.617881200244948e-05,
|
| 593 |
+
"loss": 0.0115,
|
| 594 |
+
"step": 960
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.6,
|
| 598 |
+
"learning_rate": 2.5872627066748318e-05,
|
| 599 |
+
"loss": 0.0229,
|
| 600 |
+
"step": 970
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 1.62,
|
| 604 |
+
"learning_rate": 2.5566442131047152e-05,
|
| 605 |
+
"loss": 0.0163,
|
| 606 |
+
"step": 980
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 1.64,
|
| 610 |
+
"learning_rate": 2.526025719534599e-05,
|
| 611 |
+
"loss": 0.0052,
|
| 612 |
+
"step": 990
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.65,
|
| 616 |
+
"learning_rate": 2.4954072259644827e-05,
|
| 617 |
+
"loss": 0.0237,
|
| 618 |
+
"step": 1000
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.67,
|
| 622 |
+
"learning_rate": 2.4647887323943664e-05,
|
| 623 |
+
"loss": 0.0121,
|
| 624 |
+
"step": 1010
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.69,
|
| 628 |
+
"learning_rate": 2.4341702388242498e-05,
|
| 629 |
+
"loss": 0.0022,
|
| 630 |
+
"step": 1020
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 1.7,
|
| 634 |
+
"learning_rate": 2.4035517452541335e-05,
|
| 635 |
+
"loss": 0.0105,
|
| 636 |
+
"step": 1030
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 1.72,
|
| 640 |
+
"learning_rate": 2.3729332516840173e-05,
|
| 641 |
+
"loss": 0.0138,
|
| 642 |
+
"step": 1040
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 1.74,
|
| 646 |
+
"learning_rate": 2.342314758113901e-05,
|
| 647 |
+
"loss": 0.0019,
|
| 648 |
+
"step": 1050
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.75,
|
| 652 |
+
"learning_rate": 2.3116962645437844e-05,
|
| 653 |
+
"loss": 0.0075,
|
| 654 |
+
"step": 1060
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.77,
|
| 658 |
+
"learning_rate": 2.2810777709736685e-05,
|
| 659 |
+
"loss": 0.0103,
|
| 660 |
+
"step": 1070
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.79,
|
| 664 |
+
"learning_rate": 2.250459277403552e-05,
|
| 665 |
+
"loss": 0.0107,
|
| 666 |
+
"step": 1080
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 1.8,
|
| 670 |
+
"learning_rate": 2.2198407838334356e-05,
|
| 671 |
+
"loss": 0.0018,
|
| 672 |
+
"step": 1090
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 1.82,
|
| 676 |
+
"learning_rate": 2.1892222902633193e-05,
|
| 677 |
+
"loss": 0.0038,
|
| 678 |
+
"step": 1100
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 1.83,
|
| 682 |
+
"learning_rate": 2.1586037966932027e-05,
|
| 683 |
+
"loss": 0.0028,
|
| 684 |
+
"step": 1110
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.85,
|
| 688 |
+
"learning_rate": 2.1279853031230864e-05,
|
| 689 |
+
"loss": 0.0095,
|
| 690 |
+
"step": 1120
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.87,
|
| 694 |
+
"learning_rate": 2.09736680955297e-05,
|
| 695 |
+
"loss": 0.0106,
|
| 696 |
+
"step": 1130
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.88,
|
| 700 |
+
"learning_rate": 2.066748315982854e-05,
|
| 701 |
+
"loss": 0.0285,
|
| 702 |
+
"step": 1140
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.9,
|
| 706 |
+
"learning_rate": 2.0361298224127373e-05,
|
| 707 |
+
"loss": 0.017,
|
| 708 |
+
"step": 1150
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 1.92,
|
| 712 |
+
"learning_rate": 2.005511328842621e-05,
|
| 713 |
+
"loss": 0.0025,
|
| 714 |
+
"step": 1160
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.93,
|
| 718 |
+
"learning_rate": 1.9748928352725048e-05,
|
| 719 |
+
"loss": 0.0079,
|
| 720 |
+
"step": 1170
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.95,
|
| 724 |
+
"learning_rate": 1.9442743417023885e-05,
|
| 725 |
+
"loss": 0.0167,
|
| 726 |
+
"step": 1180
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 1.97,
|
| 730 |
+
"learning_rate": 1.913655848132272e-05,
|
| 731 |
+
"loss": 0.0202,
|
| 732 |
+
"step": 1190
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.98,
|
| 736 |
+
"learning_rate": 1.8830373545621556e-05,
|
| 737 |
+
"loss": 0.0053,
|
| 738 |
+
"step": 1200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 2.0,
|
| 742 |
+
"learning_rate": 1.8524188609920394e-05,
|
| 743 |
+
"loss": 0.0079,
|
| 744 |
+
"step": 1210
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.0,
|
| 748 |
+
"eval_accuracy": 0.9980530178223753,
|
| 749 |
+
"eval_loss": 0.00832486990839243,
|
| 750 |
+
"eval_runtime": 254.4173,
|
| 751 |
+
"eval_samples_per_second": 52.489,
|
| 752 |
+
"eval_steps_per_second": 3.282,
|
| 753 |
+
"step": 1210
|
| 754 |
+
}
|
| 755 |
+
],
|
| 756 |
+
"logging_steps": 10,
|
| 757 |
+
"max_steps": 1815,
|
| 758 |
+
"num_input_tokens_seen": 0,
|
| 759 |
+
"num_train_epochs": 3,
|
| 760 |
+
"save_steps": 500,
|
| 761 |
+
"total_flos": 6.066296372278518e+18,
|
| 762 |
+
"train_batch_size": 16,
|
| 763 |
+
"trial_name": null,
|
| 764 |
+
"trial_params": null
|
| 765 |
+
}
|
research/ai_detector_v2/checkpoint-1210/training_args.bin
ADDED
|
Binary file (4.66 kB). View file
|
|
|
research/ai_detector_v2/checkpoint-1815/config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "umm-maybe/AI-image-detector",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SwinForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"depths": [
|
| 8 |
+
2,
|
| 9 |
+
2,
|
| 10 |
+
18,
|
| 11 |
+
2
|
| 12 |
+
],
|
| 13 |
+
"drop_path_rate": 0.1,
|
| 14 |
+
"embed_dim": 128,
|
| 15 |
+
"encoder_stride": 32,
|
| 16 |
+
"hidden_act": "gelu",
|
| 17 |
+
"hidden_dropout_prob": 0.0,
|
| 18 |
+
"hidden_size": 1024,
|
| 19 |
+
"id2label": {
|
| 20 |
+
"0": "ai_gen",
|
| 21 |
+
"1": "human"
|
| 22 |
+
},
|
| 23 |
+
"image_size": 224,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"label2id": {
|
| 26 |
+
"ai_gen": "0",
|
| 27 |
+
"human": "1"
|
| 28 |
+
},
|
| 29 |
+
"layer_norm_eps": 1e-05,
|
| 30 |
+
"max_length": 128,
|
| 31 |
+
"mlp_ratio": 4.0,
|
| 32 |
+
"model_type": "swin",
|
| 33 |
+
"num_channels": 3,
|
| 34 |
+
"num_heads": [
|
| 35 |
+
4,
|
| 36 |
+
8,
|
| 37 |
+
16,
|
| 38 |
+
32
|
| 39 |
+
],
|
| 40 |
+
"num_layers": 4,
|
| 41 |
+
"out_features": [
|
| 42 |
+
"stage4"
|
| 43 |
+
],
|
| 44 |
+
"out_indices": [
|
| 45 |
+
4
|
| 46 |
+
],
|
| 47 |
+
"padding": "max_length",
|
| 48 |
+
"patch_size": 4,
|
| 49 |
+
"path_norm": true,
|
| 50 |
+
"problem_type": "single_label_classification",
|
| 51 |
+
"qkv_bias": true,
|
| 52 |
+
"stage_names": [
|
| 53 |
+
"stem",
|
| 54 |
+
"stage1",
|
| 55 |
+
"stage2",
|
| 56 |
+
"stage3",
|
| 57 |
+
"stage4"
|
| 58 |
+
],
|
| 59 |
+
"torch_dtype": "float32",
|
| 60 |
+
"transformers_version": "4.37.2",
|
| 61 |
+
"use_absolute_embeddings": false,
|
| 62 |
+
"window_size": 7
|
| 63 |
+
}
|
research/ai_detector_v2/checkpoint-1815/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a9cfc3eefb0ba85fbd001d0b0e6dad70c75cae6afb2f34440f8d0916c58ba262
|
| 3 |
+
size 347498816
|
research/ai_detector_v2/checkpoint-1815/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe1835eb1c6220260837f7b7b350aabef912d71c9ca332eb1b4ae0481d592f5f
|
| 3 |
+
size 694317645
|
research/ai_detector_v2/checkpoint-1815/preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
research/ai_detector_v2/checkpoint-1815/rng_state.pth
ADDED
|
Binary file (14.2 kB). View file
|
|
|
research/ai_detector_v2/checkpoint-1815/scheduler.pt
ADDED
|
Binary file (1.06 kB). View file
|
|
|
research/ai_detector_v2/checkpoint-1815/trainer_state.json
ADDED
|
@@ -0,0 +1,1134 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9982776696121013,
|
| 3 |
+
"best_model_checkpoint": "ai_detector_v2/checkpoint-1815",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1815,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 2.747252747252747e-06,
|
| 14 |
+
"loss": 0.7391,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.03,
|
| 19 |
+
"learning_rate": 5.494505494505494e-06,
|
| 20 |
+
"loss": 0.4175,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.05,
|
| 25 |
+
"learning_rate": 8.241758241758243e-06,
|
| 26 |
+
"loss": 0.3476,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.07,
|
| 31 |
+
"learning_rate": 1.0989010989010989e-05,
|
| 32 |
+
"loss": 0.1961,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.08,
|
| 37 |
+
"learning_rate": 1.3736263736263738e-05,
|
| 38 |
+
"loss": 0.1436,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.1,
|
| 43 |
+
"learning_rate": 1.6483516483516486e-05,
|
| 44 |
+
"loss": 0.0595,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.12,
|
| 49 |
+
"learning_rate": 1.923076923076923e-05,
|
| 50 |
+
"loss": 0.0585,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.13,
|
| 55 |
+
"learning_rate": 2.1978021978021977e-05,
|
| 56 |
+
"loss": 0.0607,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.15,
|
| 61 |
+
"learning_rate": 2.4725274725274727e-05,
|
| 62 |
+
"loss": 0.041,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.17,
|
| 67 |
+
"learning_rate": 2.7472527472527476e-05,
|
| 68 |
+
"loss": 0.0561,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.18,
|
| 73 |
+
"learning_rate": 3.021978021978022e-05,
|
| 74 |
+
"loss": 0.0517,
|
| 75 |
+
"step": 110
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.2,
|
| 79 |
+
"learning_rate": 3.296703296703297e-05,
|
| 80 |
+
"loss": 0.0581,
|
| 81 |
+
"step": 120
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.21,
|
| 85 |
+
"learning_rate": 3.571428571428572e-05,
|
| 86 |
+
"loss": 0.0281,
|
| 87 |
+
"step": 130
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.23,
|
| 91 |
+
"learning_rate": 3.846153846153846e-05,
|
| 92 |
+
"loss": 0.0612,
|
| 93 |
+
"step": 140
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.25,
|
| 97 |
+
"learning_rate": 4.120879120879121e-05,
|
| 98 |
+
"loss": 0.0297,
|
| 99 |
+
"step": 150
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.26,
|
| 103 |
+
"learning_rate": 4.3956043956043955e-05,
|
| 104 |
+
"loss": 0.021,
|
| 105 |
+
"step": 160
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.28,
|
| 109 |
+
"learning_rate": 4.670329670329671e-05,
|
| 110 |
+
"loss": 0.019,
|
| 111 |
+
"step": 170
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.3,
|
| 115 |
+
"learning_rate": 4.945054945054945e-05,
|
| 116 |
+
"loss": 0.0233,
|
| 117 |
+
"step": 180
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.31,
|
| 121 |
+
"learning_rate": 4.9755052051439074e-05,
|
| 122 |
+
"loss": 0.0298,
|
| 123 |
+
"step": 190
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.33,
|
| 127 |
+
"learning_rate": 4.944886711573791e-05,
|
| 128 |
+
"loss": 0.0421,
|
| 129 |
+
"step": 200
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.35,
|
| 133 |
+
"learning_rate": 4.914268218003674e-05,
|
| 134 |
+
"loss": 0.0277,
|
| 135 |
+
"step": 210
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.36,
|
| 139 |
+
"learning_rate": 4.883649724433558e-05,
|
| 140 |
+
"loss": 0.0361,
|
| 141 |
+
"step": 220
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.38,
|
| 145 |
+
"learning_rate": 4.8530312308634416e-05,
|
| 146 |
+
"loss": 0.0579,
|
| 147 |
+
"step": 230
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.4,
|
| 151 |
+
"learning_rate": 4.822412737293325e-05,
|
| 152 |
+
"loss": 0.0338,
|
| 153 |
+
"step": 240
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.41,
|
| 157 |
+
"learning_rate": 4.791794243723209e-05,
|
| 158 |
+
"loss": 0.0366,
|
| 159 |
+
"step": 250
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.43,
|
| 163 |
+
"learning_rate": 4.761175750153093e-05,
|
| 164 |
+
"loss": 0.0092,
|
| 165 |
+
"step": 260
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.45,
|
| 169 |
+
"learning_rate": 4.7305572565829766e-05,
|
| 170 |
+
"loss": 0.063,
|
| 171 |
+
"step": 270
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.46,
|
| 175 |
+
"learning_rate": 4.69993876301286e-05,
|
| 176 |
+
"loss": 0.0527,
|
| 177 |
+
"step": 280
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.48,
|
| 181 |
+
"learning_rate": 4.6693202694427433e-05,
|
| 182 |
+
"loss": 0.023,
|
| 183 |
+
"step": 290
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.5,
|
| 187 |
+
"learning_rate": 4.6387017758726274e-05,
|
| 188 |
+
"loss": 0.0437,
|
| 189 |
+
"step": 300
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.51,
|
| 193 |
+
"learning_rate": 4.608083282302511e-05,
|
| 194 |
+
"loss": 0.0381,
|
| 195 |
+
"step": 310
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.53,
|
| 199 |
+
"learning_rate": 4.577464788732395e-05,
|
| 200 |
+
"loss": 0.0288,
|
| 201 |
+
"step": 320
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.55,
|
| 205 |
+
"learning_rate": 4.546846295162278e-05,
|
| 206 |
+
"loss": 0.0385,
|
| 207 |
+
"step": 330
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.56,
|
| 211 |
+
"learning_rate": 4.516227801592162e-05,
|
| 212 |
+
"loss": 0.0442,
|
| 213 |
+
"step": 340
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.58,
|
| 217 |
+
"learning_rate": 4.485609308022046e-05,
|
| 218 |
+
"loss": 0.0128,
|
| 219 |
+
"step": 350
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6,
|
| 223 |
+
"learning_rate": 4.454990814451929e-05,
|
| 224 |
+
"loss": 0.0189,
|
| 225 |
+
"step": 360
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.61,
|
| 229 |
+
"learning_rate": 4.4243723208818125e-05,
|
| 230 |
+
"loss": 0.0219,
|
| 231 |
+
"step": 370
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.63,
|
| 235 |
+
"learning_rate": 4.3937538273116966e-05,
|
| 236 |
+
"loss": 0.0296,
|
| 237 |
+
"step": 380
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.64,
|
| 241 |
+
"learning_rate": 4.363135333741581e-05,
|
| 242 |
+
"loss": 0.058,
|
| 243 |
+
"step": 390
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.66,
|
| 247 |
+
"learning_rate": 4.332516840171464e-05,
|
| 248 |
+
"loss": 0.0317,
|
| 249 |
+
"step": 400
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.68,
|
| 253 |
+
"learning_rate": 4.3018983466013475e-05,
|
| 254 |
+
"loss": 0.0364,
|
| 255 |
+
"step": 410
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.69,
|
| 259 |
+
"learning_rate": 4.271279853031231e-05,
|
| 260 |
+
"loss": 0.0239,
|
| 261 |
+
"step": 420
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.71,
|
| 265 |
+
"learning_rate": 4.240661359461115e-05,
|
| 266 |
+
"loss": 0.0265,
|
| 267 |
+
"step": 430
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.73,
|
| 271 |
+
"learning_rate": 4.210042865890998e-05,
|
| 272 |
+
"loss": 0.0046,
|
| 273 |
+
"step": 440
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.74,
|
| 277 |
+
"learning_rate": 4.179424372320882e-05,
|
| 278 |
+
"loss": 0.0182,
|
| 279 |
+
"step": 450
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.76,
|
| 283 |
+
"learning_rate": 4.148805878750766e-05,
|
| 284 |
+
"loss": 0.0255,
|
| 285 |
+
"step": 460
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.78,
|
| 289 |
+
"learning_rate": 4.118187385180649e-05,
|
| 290 |
+
"loss": 0.0143,
|
| 291 |
+
"step": 470
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.79,
|
| 295 |
+
"learning_rate": 4.087568891610533e-05,
|
| 296 |
+
"loss": 0.0348,
|
| 297 |
+
"step": 480
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.81,
|
| 301 |
+
"learning_rate": 4.0569503980404167e-05,
|
| 302 |
+
"loss": 0.035,
|
| 303 |
+
"step": 490
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.83,
|
| 307 |
+
"learning_rate": 4.0263319044703e-05,
|
| 308 |
+
"loss": 0.0397,
|
| 309 |
+
"step": 500
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.84,
|
| 313 |
+
"learning_rate": 3.9957134109001834e-05,
|
| 314 |
+
"loss": 0.0194,
|
| 315 |
+
"step": 510
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.86,
|
| 319 |
+
"learning_rate": 3.9650949173300675e-05,
|
| 320 |
+
"loss": 0.0075,
|
| 321 |
+
"step": 520
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.88,
|
| 325 |
+
"learning_rate": 3.9344764237599516e-05,
|
| 326 |
+
"loss": 0.0025,
|
| 327 |
+
"step": 530
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.89,
|
| 331 |
+
"learning_rate": 3.903857930189835e-05,
|
| 332 |
+
"loss": 0.002,
|
| 333 |
+
"step": 540
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.91,
|
| 337 |
+
"learning_rate": 3.8732394366197184e-05,
|
| 338 |
+
"loss": 0.0093,
|
| 339 |
+
"step": 550
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.93,
|
| 343 |
+
"learning_rate": 3.8426209430496025e-05,
|
| 344 |
+
"loss": 0.0027,
|
| 345 |
+
"step": 560
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.94,
|
| 349 |
+
"learning_rate": 3.812002449479486e-05,
|
| 350 |
+
"loss": 0.011,
|
| 351 |
+
"step": 570
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 0.96,
|
| 355 |
+
"learning_rate": 3.781383955909369e-05,
|
| 356 |
+
"loss": 0.0133,
|
| 357 |
+
"step": 580
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.98,
|
| 361 |
+
"learning_rate": 3.7507654623392526e-05,
|
| 362 |
+
"loss": 0.0314,
|
| 363 |
+
"step": 590
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.99,
|
| 367 |
+
"learning_rate": 3.720146968769137e-05,
|
| 368 |
+
"loss": 0.0187,
|
| 369 |
+
"step": 600
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.0,
|
| 373 |
+
"eval_accuracy": 0.9973790624531975,
|
| 374 |
+
"eval_loss": 0.010242484509944916,
|
| 375 |
+
"eval_runtime": 251.9013,
|
| 376 |
+
"eval_samples_per_second": 53.013,
|
| 377 |
+
"eval_steps_per_second": 3.315,
|
| 378 |
+
"step": 605
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.01,
|
| 382 |
+
"learning_rate": 3.689528475199021e-05,
|
| 383 |
+
"loss": 0.0335,
|
| 384 |
+
"step": 610
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.02,
|
| 388 |
+
"learning_rate": 3.658909981628904e-05,
|
| 389 |
+
"loss": 0.0224,
|
| 390 |
+
"step": 620
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.04,
|
| 394 |
+
"learning_rate": 3.6282914880587876e-05,
|
| 395 |
+
"loss": 0.0192,
|
| 396 |
+
"step": 630
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.06,
|
| 400 |
+
"learning_rate": 3.597672994488671e-05,
|
| 401 |
+
"loss": 0.0164,
|
| 402 |
+
"step": 640
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.07,
|
| 406 |
+
"learning_rate": 3.567054500918555e-05,
|
| 407 |
+
"loss": 0.0109,
|
| 408 |
+
"step": 650
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.09,
|
| 412 |
+
"learning_rate": 3.5364360073484384e-05,
|
| 413 |
+
"loss": 0.013,
|
| 414 |
+
"step": 660
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.11,
|
| 418 |
+
"learning_rate": 3.5058175137783225e-05,
|
| 419 |
+
"loss": 0.0055,
|
| 420 |
+
"step": 670
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 1.12,
|
| 424 |
+
"learning_rate": 3.475199020208206e-05,
|
| 425 |
+
"loss": 0.0191,
|
| 426 |
+
"step": 680
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 1.14,
|
| 430 |
+
"learning_rate": 3.44458052663809e-05,
|
| 431 |
+
"loss": 0.0013,
|
| 432 |
+
"step": 690
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 1.16,
|
| 436 |
+
"learning_rate": 3.4139620330679734e-05,
|
| 437 |
+
"loss": 0.0186,
|
| 438 |
+
"step": 700
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 1.17,
|
| 442 |
+
"learning_rate": 3.383343539497857e-05,
|
| 443 |
+
"loss": 0.0001,
|
| 444 |
+
"step": 710
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.19,
|
| 448 |
+
"learning_rate": 3.35272504592774e-05,
|
| 449 |
+
"loss": 0.0127,
|
| 450 |
+
"step": 720
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.21,
|
| 454 |
+
"learning_rate": 3.322106552357624e-05,
|
| 455 |
+
"loss": 0.0052,
|
| 456 |
+
"step": 730
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 1.22,
|
| 460 |
+
"learning_rate": 3.291488058787508e-05,
|
| 461 |
+
"loss": 0.0278,
|
| 462 |
+
"step": 740
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 1.24,
|
| 466 |
+
"learning_rate": 3.260869565217392e-05,
|
| 467 |
+
"loss": 0.0015,
|
| 468 |
+
"step": 750
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 1.26,
|
| 472 |
+
"learning_rate": 3.230251071647275e-05,
|
| 473 |
+
"loss": 0.0026,
|
| 474 |
+
"step": 760
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.27,
|
| 478 |
+
"learning_rate": 3.1996325780771585e-05,
|
| 479 |
+
"loss": 0.0243,
|
| 480 |
+
"step": 770
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 1.29,
|
| 484 |
+
"learning_rate": 3.1690140845070426e-05,
|
| 485 |
+
"loss": 0.0142,
|
| 486 |
+
"step": 780
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.31,
|
| 490 |
+
"learning_rate": 3.138395590936926e-05,
|
| 491 |
+
"loss": 0.007,
|
| 492 |
+
"step": 790
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.32,
|
| 496 |
+
"learning_rate": 3.1077770973668093e-05,
|
| 497 |
+
"loss": 0.0175,
|
| 498 |
+
"step": 800
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 1.34,
|
| 502 |
+
"learning_rate": 3.0771586037966934e-05,
|
| 503 |
+
"loss": 0.0064,
|
| 504 |
+
"step": 810
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.36,
|
| 508 |
+
"learning_rate": 3.046540110226577e-05,
|
| 509 |
+
"loss": 0.0168,
|
| 510 |
+
"step": 820
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 1.37,
|
| 514 |
+
"learning_rate": 3.015921616656461e-05,
|
| 515 |
+
"loss": 0.002,
|
| 516 |
+
"step": 830
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 1.39,
|
| 520 |
+
"learning_rate": 2.9853031230863443e-05,
|
| 521 |
+
"loss": 0.0129,
|
| 522 |
+
"step": 840
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 1.4,
|
| 526 |
+
"learning_rate": 2.9546846295162277e-05,
|
| 527 |
+
"loss": 0.0134,
|
| 528 |
+
"step": 850
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.42,
|
| 532 |
+
"learning_rate": 2.9240661359461114e-05,
|
| 533 |
+
"loss": 0.0194,
|
| 534 |
+
"step": 860
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.44,
|
| 538 |
+
"learning_rate": 2.8934476423759955e-05,
|
| 539 |
+
"loss": 0.0096,
|
| 540 |
+
"step": 870
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 1.45,
|
| 544 |
+
"learning_rate": 2.862829148805879e-05,
|
| 545 |
+
"loss": 0.0048,
|
| 546 |
+
"step": 880
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.47,
|
| 550 |
+
"learning_rate": 2.8322106552357626e-05,
|
| 551 |
+
"loss": 0.0003,
|
| 552 |
+
"step": 890
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 1.49,
|
| 556 |
+
"learning_rate": 2.801592161665646e-05,
|
| 557 |
+
"loss": 0.0115,
|
| 558 |
+
"step": 900
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 1.5,
|
| 562 |
+
"learning_rate": 2.77097366809553e-05,
|
| 563 |
+
"loss": 0.0264,
|
| 564 |
+
"step": 910
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.52,
|
| 568 |
+
"learning_rate": 2.7403551745254135e-05,
|
| 569 |
+
"loss": 0.0233,
|
| 570 |
+
"step": 920
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.54,
|
| 574 |
+
"learning_rate": 2.7097366809552972e-05,
|
| 575 |
+
"loss": 0.0008,
|
| 576 |
+
"step": 930
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.55,
|
| 580 |
+
"learning_rate": 2.6791181873851806e-05,
|
| 581 |
+
"loss": 0.0103,
|
| 582 |
+
"step": 940
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 1.57,
|
| 586 |
+
"learning_rate": 2.6484996938150647e-05,
|
| 587 |
+
"loss": 0.0356,
|
| 588 |
+
"step": 950
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 1.59,
|
| 592 |
+
"learning_rate": 2.617881200244948e-05,
|
| 593 |
+
"loss": 0.0115,
|
| 594 |
+
"step": 960
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.6,
|
| 598 |
+
"learning_rate": 2.5872627066748318e-05,
|
| 599 |
+
"loss": 0.0229,
|
| 600 |
+
"step": 970
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 1.62,
|
| 604 |
+
"learning_rate": 2.5566442131047152e-05,
|
| 605 |
+
"loss": 0.0163,
|
| 606 |
+
"step": 980
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 1.64,
|
| 610 |
+
"learning_rate": 2.526025719534599e-05,
|
| 611 |
+
"loss": 0.0052,
|
| 612 |
+
"step": 990
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.65,
|
| 616 |
+
"learning_rate": 2.4954072259644827e-05,
|
| 617 |
+
"loss": 0.0237,
|
| 618 |
+
"step": 1000
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.67,
|
| 622 |
+
"learning_rate": 2.4647887323943664e-05,
|
| 623 |
+
"loss": 0.0121,
|
| 624 |
+
"step": 1010
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.69,
|
| 628 |
+
"learning_rate": 2.4341702388242498e-05,
|
| 629 |
+
"loss": 0.0022,
|
| 630 |
+
"step": 1020
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 1.7,
|
| 634 |
+
"learning_rate": 2.4035517452541335e-05,
|
| 635 |
+
"loss": 0.0105,
|
| 636 |
+
"step": 1030
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 1.72,
|
| 640 |
+
"learning_rate": 2.3729332516840173e-05,
|
| 641 |
+
"loss": 0.0138,
|
| 642 |
+
"step": 1040
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 1.74,
|
| 646 |
+
"learning_rate": 2.342314758113901e-05,
|
| 647 |
+
"loss": 0.0019,
|
| 648 |
+
"step": 1050
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.75,
|
| 652 |
+
"learning_rate": 2.3116962645437844e-05,
|
| 653 |
+
"loss": 0.0075,
|
| 654 |
+
"step": 1060
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.77,
|
| 658 |
+
"learning_rate": 2.2810777709736685e-05,
|
| 659 |
+
"loss": 0.0103,
|
| 660 |
+
"step": 1070
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.79,
|
| 664 |
+
"learning_rate": 2.250459277403552e-05,
|
| 665 |
+
"loss": 0.0107,
|
| 666 |
+
"step": 1080
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 1.8,
|
| 670 |
+
"learning_rate": 2.2198407838334356e-05,
|
| 671 |
+
"loss": 0.0018,
|
| 672 |
+
"step": 1090
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 1.82,
|
| 676 |
+
"learning_rate": 2.1892222902633193e-05,
|
| 677 |
+
"loss": 0.0038,
|
| 678 |
+
"step": 1100
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 1.83,
|
| 682 |
+
"learning_rate": 2.1586037966932027e-05,
|
| 683 |
+
"loss": 0.0028,
|
| 684 |
+
"step": 1110
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.85,
|
| 688 |
+
"learning_rate": 2.1279853031230864e-05,
|
| 689 |
+
"loss": 0.0095,
|
| 690 |
+
"step": 1120
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.87,
|
| 694 |
+
"learning_rate": 2.09736680955297e-05,
|
| 695 |
+
"loss": 0.0106,
|
| 696 |
+
"step": 1130
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.88,
|
| 700 |
+
"learning_rate": 2.066748315982854e-05,
|
| 701 |
+
"loss": 0.0285,
|
| 702 |
+
"step": 1140
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.9,
|
| 706 |
+
"learning_rate": 2.0361298224127373e-05,
|
| 707 |
+
"loss": 0.017,
|
| 708 |
+
"step": 1150
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 1.92,
|
| 712 |
+
"learning_rate": 2.005511328842621e-05,
|
| 713 |
+
"loss": 0.0025,
|
| 714 |
+
"step": 1160
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.93,
|
| 718 |
+
"learning_rate": 1.9748928352725048e-05,
|
| 719 |
+
"loss": 0.0079,
|
| 720 |
+
"step": 1170
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.95,
|
| 724 |
+
"learning_rate": 1.9442743417023885e-05,
|
| 725 |
+
"loss": 0.0167,
|
| 726 |
+
"step": 1180
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 1.97,
|
| 730 |
+
"learning_rate": 1.913655848132272e-05,
|
| 731 |
+
"loss": 0.0202,
|
| 732 |
+
"step": 1190
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.98,
|
| 736 |
+
"learning_rate": 1.8830373545621556e-05,
|
| 737 |
+
"loss": 0.0053,
|
| 738 |
+
"step": 1200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 2.0,
|
| 742 |
+
"learning_rate": 1.8524188609920394e-05,
|
| 743 |
+
"loss": 0.0079,
|
| 744 |
+
"step": 1210
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.0,
|
| 748 |
+
"eval_accuracy": 0.9980530178223753,
|
| 749 |
+
"eval_loss": 0.00832486990839243,
|
| 750 |
+
"eval_runtime": 254.4173,
|
| 751 |
+
"eval_samples_per_second": 52.489,
|
| 752 |
+
"eval_steps_per_second": 3.282,
|
| 753 |
+
"step": 1210
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 2.02,
|
| 757 |
+
"learning_rate": 1.821800367421923e-05,
|
| 758 |
+
"loss": 0.0127,
|
| 759 |
+
"step": 1220
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 2.03,
|
| 763 |
+
"learning_rate": 1.7911818738518065e-05,
|
| 764 |
+
"loss": 0.0273,
|
| 765 |
+
"step": 1230
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.05,
|
| 769 |
+
"learning_rate": 1.7605633802816902e-05,
|
| 770 |
+
"loss": 0.0085,
|
| 771 |
+
"step": 1240
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 2.07,
|
| 775 |
+
"learning_rate": 1.729944886711574e-05,
|
| 776 |
+
"loss": 0.0045,
|
| 777 |
+
"step": 1250
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 2.08,
|
| 781 |
+
"learning_rate": 1.6993263931414574e-05,
|
| 782 |
+
"loss": 0.0048,
|
| 783 |
+
"step": 1260
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 2.1,
|
| 787 |
+
"learning_rate": 1.668707899571341e-05,
|
| 788 |
+
"loss": 0.0016,
|
| 789 |
+
"step": 1270
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 2.12,
|
| 793 |
+
"learning_rate": 1.6380894060012248e-05,
|
| 794 |
+
"loss": 0.0012,
|
| 795 |
+
"step": 1280
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 2.13,
|
| 799 |
+
"learning_rate": 1.6074709124311086e-05,
|
| 800 |
+
"loss": 0.0065,
|
| 801 |
+
"step": 1290
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 2.15,
|
| 805 |
+
"learning_rate": 1.576852418860992e-05,
|
| 806 |
+
"loss": 0.005,
|
| 807 |
+
"step": 1300
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.17,
|
| 811 |
+
"learning_rate": 1.5462339252908757e-05,
|
| 812 |
+
"loss": 0.0421,
|
| 813 |
+
"step": 1310
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.18,
|
| 817 |
+
"learning_rate": 1.5156154317207594e-05,
|
| 818 |
+
"loss": 0.0077,
|
| 819 |
+
"step": 1320
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 2.2,
|
| 823 |
+
"learning_rate": 1.4849969381506432e-05,
|
| 824 |
+
"loss": 0.0012,
|
| 825 |
+
"step": 1330
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 2.21,
|
| 829 |
+
"learning_rate": 1.4543784445805267e-05,
|
| 830 |
+
"loss": 0.0091,
|
| 831 |
+
"step": 1340
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 2.23,
|
| 835 |
+
"learning_rate": 1.4237599510104105e-05,
|
| 836 |
+
"loss": 0.0091,
|
| 837 |
+
"step": 1350
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 2.25,
|
| 841 |
+
"learning_rate": 1.393141457440294e-05,
|
| 842 |
+
"loss": 0.0256,
|
| 843 |
+
"step": 1360
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 2.26,
|
| 847 |
+
"learning_rate": 1.3625229638701776e-05,
|
| 848 |
+
"loss": 0.0114,
|
| 849 |
+
"step": 1370
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.28,
|
| 853 |
+
"learning_rate": 1.3319044703000613e-05,
|
| 854 |
+
"loss": 0.0059,
|
| 855 |
+
"step": 1380
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 2.3,
|
| 859 |
+
"learning_rate": 1.3012859767299449e-05,
|
| 860 |
+
"loss": 0.0012,
|
| 861 |
+
"step": 1390
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 2.31,
|
| 865 |
+
"learning_rate": 1.2706674831598286e-05,
|
| 866 |
+
"loss": 0.0031,
|
| 867 |
+
"step": 1400
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 2.33,
|
| 871 |
+
"learning_rate": 1.2400489895897123e-05,
|
| 872 |
+
"loss": 0.0005,
|
| 873 |
+
"step": 1410
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 2.35,
|
| 877 |
+
"learning_rate": 1.2094304960195959e-05,
|
| 878 |
+
"loss": 0.0013,
|
| 879 |
+
"step": 1420
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 2.36,
|
| 883 |
+
"learning_rate": 1.1788120024494795e-05,
|
| 884 |
+
"loss": 0.0006,
|
| 885 |
+
"step": 1430
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 2.38,
|
| 889 |
+
"learning_rate": 1.1481935088793632e-05,
|
| 890 |
+
"loss": 0.011,
|
| 891 |
+
"step": 1440
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.4,
|
| 895 |
+
"learning_rate": 1.1175750153092468e-05,
|
| 896 |
+
"loss": 0.0173,
|
| 897 |
+
"step": 1450
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 2.41,
|
| 901 |
+
"learning_rate": 1.0869565217391305e-05,
|
| 902 |
+
"loss": 0.0077,
|
| 903 |
+
"step": 1460
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 2.43,
|
| 907 |
+
"learning_rate": 1.056338028169014e-05,
|
| 908 |
+
"loss": 0.0007,
|
| 909 |
+
"step": 1470
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 2.45,
|
| 913 |
+
"learning_rate": 1.0257195345988978e-05,
|
| 914 |
+
"loss": 0.0077,
|
| 915 |
+
"step": 1480
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 2.46,
|
| 919 |
+
"learning_rate": 9.951010410287814e-06,
|
| 920 |
+
"loss": 0.0093,
|
| 921 |
+
"step": 1490
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 2.48,
|
| 925 |
+
"learning_rate": 9.644825474586651e-06,
|
| 926 |
+
"loss": 0.01,
|
| 927 |
+
"step": 1500
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 2.5,
|
| 931 |
+
"learning_rate": 9.338640538885487e-06,
|
| 932 |
+
"loss": 0.0016,
|
| 933 |
+
"step": 1510
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.51,
|
| 937 |
+
"learning_rate": 9.032455603184324e-06,
|
| 938 |
+
"loss": 0.0001,
|
| 939 |
+
"step": 1520
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 2.53,
|
| 943 |
+
"learning_rate": 8.726270667483161e-06,
|
| 944 |
+
"loss": 0.0068,
|
| 945 |
+
"step": 1530
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 2.55,
|
| 949 |
+
"learning_rate": 8.420085731781997e-06,
|
| 950 |
+
"loss": 0.0007,
|
| 951 |
+
"step": 1540
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 2.56,
|
| 955 |
+
"learning_rate": 8.113900796080834e-06,
|
| 956 |
+
"loss": 0.0093,
|
| 957 |
+
"step": 1550
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 2.58,
|
| 961 |
+
"learning_rate": 7.807715860379668e-06,
|
| 962 |
+
"loss": 0.0004,
|
| 963 |
+
"step": 1560
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 2.6,
|
| 967 |
+
"learning_rate": 7.501530924678506e-06,
|
| 968 |
+
"loss": 0.0051,
|
| 969 |
+
"step": 1570
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 2.61,
|
| 973 |
+
"learning_rate": 7.195345988977342e-06,
|
| 974 |
+
"loss": 0.0046,
|
| 975 |
+
"step": 1580
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.63,
|
| 979 |
+
"learning_rate": 6.889161053276179e-06,
|
| 980 |
+
"loss": 0.0002,
|
| 981 |
+
"step": 1590
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 2.64,
|
| 985 |
+
"learning_rate": 6.582976117575015e-06,
|
| 986 |
+
"loss": 0.0059,
|
| 987 |
+
"step": 1600
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 2.66,
|
| 991 |
+
"learning_rate": 6.276791181873852e-06,
|
| 992 |
+
"loss": 0.0038,
|
| 993 |
+
"step": 1610
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 2.68,
|
| 997 |
+
"learning_rate": 5.970606246172689e-06,
|
| 998 |
+
"loss": 0.0002,
|
| 999 |
+
"step": 1620
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.69,
|
| 1003 |
+
"learning_rate": 5.664421310471525e-06,
|
| 1004 |
+
"loss": 0.0003,
|
| 1005 |
+
"step": 1630
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 2.71,
|
| 1009 |
+
"learning_rate": 5.358236374770362e-06,
|
| 1010 |
+
"loss": 0.0048,
|
| 1011 |
+
"step": 1640
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 2.73,
|
| 1015 |
+
"learning_rate": 5.052051439069198e-06,
|
| 1016 |
+
"loss": 0.001,
|
| 1017 |
+
"step": 1650
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.74,
|
| 1021 |
+
"learning_rate": 4.745866503368034e-06,
|
| 1022 |
+
"loss": 0.0006,
|
| 1023 |
+
"step": 1660
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 2.76,
|
| 1027 |
+
"learning_rate": 4.4396815676668705e-06,
|
| 1028 |
+
"loss": 0.012,
|
| 1029 |
+
"step": 1670
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 2.78,
|
| 1033 |
+
"learning_rate": 4.133496631965708e-06,
|
| 1034 |
+
"loss": 0.0002,
|
| 1035 |
+
"step": 1680
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 2.79,
|
| 1039 |
+
"learning_rate": 3.827311696264544e-06,
|
| 1040 |
+
"loss": 0.0001,
|
| 1041 |
+
"step": 1690
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 2.81,
|
| 1045 |
+
"learning_rate": 3.521126760563381e-06,
|
| 1046 |
+
"loss": 0.0026,
|
| 1047 |
+
"step": 1700
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 2.83,
|
| 1051 |
+
"learning_rate": 3.2149418248622173e-06,
|
| 1052 |
+
"loss": 0.0085,
|
| 1053 |
+
"step": 1710
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 2.84,
|
| 1057 |
+
"learning_rate": 2.9087568891610534e-06,
|
| 1058 |
+
"loss": 0.0096,
|
| 1059 |
+
"step": 1720
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.86,
|
| 1063 |
+
"learning_rate": 2.60257195345989e-06,
|
| 1064 |
+
"loss": 0.0001,
|
| 1065 |
+
"step": 1730
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 2.88,
|
| 1069 |
+
"learning_rate": 2.2963870177587263e-06,
|
| 1070 |
+
"loss": 0.0076,
|
| 1071 |
+
"step": 1740
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 2.89,
|
| 1075 |
+
"learning_rate": 1.990202082057563e-06,
|
| 1076 |
+
"loss": 0.0002,
|
| 1077 |
+
"step": 1750
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 2.91,
|
| 1081 |
+
"learning_rate": 1.6840171463563995e-06,
|
| 1082 |
+
"loss": 0.0044,
|
| 1083 |
+
"step": 1760
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 2.93,
|
| 1087 |
+
"learning_rate": 1.3778322106552358e-06,
|
| 1088 |
+
"loss": 0.0005,
|
| 1089 |
+
"step": 1770
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 2.94,
|
| 1093 |
+
"learning_rate": 1.0716472749540723e-06,
|
| 1094 |
+
"loss": 0.0001,
|
| 1095 |
+
"step": 1780
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 2.96,
|
| 1099 |
+
"learning_rate": 7.654623392529088e-07,
|
| 1100 |
+
"loss": 0.0024,
|
| 1101 |
+
"step": 1790
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.98,
|
| 1105 |
+
"learning_rate": 4.5927740355174527e-07,
|
| 1106 |
+
"loss": 0.0001,
|
| 1107 |
+
"step": 1800
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 2.99,
|
| 1111 |
+
"learning_rate": 1.5309246785058176e-07,
|
| 1112 |
+
"loss": 0.0002,
|
| 1113 |
+
"step": 1810
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 3.0,
|
| 1117 |
+
"eval_accuracy": 0.9982776696121013,
|
| 1118 |
+
"eval_loss": 0.007270948030054569,
|
| 1119 |
+
"eval_runtime": 259.3731,
|
| 1120 |
+
"eval_samples_per_second": 51.486,
|
| 1121 |
+
"eval_steps_per_second": 3.219,
|
| 1122 |
+
"step": 1815
|
| 1123 |
+
}
|
| 1124 |
+
],
|
| 1125 |
+
"logging_steps": 10,
|
| 1126 |
+
"max_steps": 1815,
|
| 1127 |
+
"num_input_tokens_seen": 0,
|
| 1128 |
+
"num_train_epochs": 3,
|
| 1129 |
+
"save_steps": 500,
|
| 1130 |
+
"total_flos": 9.099444558417777e+18,
|
| 1131 |
+
"train_batch_size": 16,
|
| 1132 |
+
"trial_name": null,
|
| 1133 |
+
"trial_params": null
|
| 1134 |
+
}
|
research/ai_detector_v2/checkpoint-1815/training_args.bin
ADDED
|
Binary file (4.66 kB). View file
|
|
|
research/ai_detector_v2/checkpoint-605/config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "umm-maybe/AI-image-detector",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SwinForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"depths": [
|
| 8 |
+
2,
|
| 9 |
+
2,
|
| 10 |
+
18,
|
| 11 |
+
2
|
| 12 |
+
],
|
| 13 |
+
"drop_path_rate": 0.1,
|
| 14 |
+
"embed_dim": 128,
|
| 15 |
+
"encoder_stride": 32,
|
| 16 |
+
"hidden_act": "gelu",
|
| 17 |
+
"hidden_dropout_prob": 0.0,
|
| 18 |
+
"hidden_size": 1024,
|
| 19 |
+
"id2label": {
|
| 20 |
+
"0": "ai_gen",
|
| 21 |
+
"1": "human"
|
| 22 |
+
},
|
| 23 |
+
"image_size": 224,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"label2id": {
|
| 26 |
+
"ai_gen": "0",
|
| 27 |
+
"human": "1"
|
| 28 |
+
},
|
| 29 |
+
"layer_norm_eps": 1e-05,
|
| 30 |
+
"max_length": 128,
|
| 31 |
+
"mlp_ratio": 4.0,
|
| 32 |
+
"model_type": "swin",
|
| 33 |
+
"num_channels": 3,
|
| 34 |
+
"num_heads": [
|
| 35 |
+
4,
|
| 36 |
+
8,
|
| 37 |
+
16,
|
| 38 |
+
32
|
| 39 |
+
],
|
| 40 |
+
"num_layers": 4,
|
| 41 |
+
"out_features": [
|
| 42 |
+
"stage4"
|
| 43 |
+
],
|
| 44 |
+
"out_indices": [
|
| 45 |
+
4
|
| 46 |
+
],
|
| 47 |
+
"padding": "max_length",
|
| 48 |
+
"patch_size": 4,
|
| 49 |
+
"path_norm": true,
|
| 50 |
+
"problem_type": "single_label_classification",
|
| 51 |
+
"qkv_bias": true,
|
| 52 |
+
"stage_names": [
|
| 53 |
+
"stem",
|
| 54 |
+
"stage1",
|
| 55 |
+
"stage2",
|
| 56 |
+
"stage3",
|
| 57 |
+
"stage4"
|
| 58 |
+
],
|
| 59 |
+
"torch_dtype": "float32",
|
| 60 |
+
"transformers_version": "4.37.2",
|
| 61 |
+
"use_absolute_embeddings": false,
|
| 62 |
+
"window_size": 7
|
| 63 |
+
}
|
research/ai_detector_v2/checkpoint-605/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a29b82064c6e5bf3f5657a83c36e71b62d7a33a333123f90919c243b2285ed89
|
| 3 |
+
size 347498816
|
research/ai_detector_v2/checkpoint-605/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7926d08708895df2e9db701575d1770ec5a9941783292a0ff61156f4297a9dd8
|
| 3 |
+
size 694317645
|
research/ai_detector_v2/checkpoint-605/preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
research/ai_detector_v2/checkpoint-605/rng_state.pth
ADDED
|
Binary file (14.2 kB). View file
|
|
|
research/ai_detector_v2/checkpoint-605/scheduler.pt
ADDED
|
Binary file (1.06 kB). View file
|
|
|
research/ai_detector_v2/checkpoint-605/trainer_state.json
ADDED
|
@@ -0,0 +1,390 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9973790624531975,
|
| 3 |
+
"best_model_checkpoint": "ai_detector_v2/checkpoint-605",
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 605,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 2.747252747252747e-06,
|
| 14 |
+
"loss": 0.7391,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.03,
|
| 19 |
+
"learning_rate": 5.494505494505494e-06,
|
| 20 |
+
"loss": 0.4175,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.05,
|
| 25 |
+
"learning_rate": 8.241758241758243e-06,
|
| 26 |
+
"loss": 0.3476,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.07,
|
| 31 |
+
"learning_rate": 1.0989010989010989e-05,
|
| 32 |
+
"loss": 0.1961,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.08,
|
| 37 |
+
"learning_rate": 1.3736263736263738e-05,
|
| 38 |
+
"loss": 0.1436,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.1,
|
| 43 |
+
"learning_rate": 1.6483516483516486e-05,
|
| 44 |
+
"loss": 0.0595,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.12,
|
| 49 |
+
"learning_rate": 1.923076923076923e-05,
|
| 50 |
+
"loss": 0.0585,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.13,
|
| 55 |
+
"learning_rate": 2.1978021978021977e-05,
|
| 56 |
+
"loss": 0.0607,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.15,
|
| 61 |
+
"learning_rate": 2.4725274725274727e-05,
|
| 62 |
+
"loss": 0.041,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.17,
|
| 67 |
+
"learning_rate": 2.7472527472527476e-05,
|
| 68 |
+
"loss": 0.0561,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.18,
|
| 73 |
+
"learning_rate": 3.021978021978022e-05,
|
| 74 |
+
"loss": 0.0517,
|
| 75 |
+
"step": 110
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.2,
|
| 79 |
+
"learning_rate": 3.296703296703297e-05,
|
| 80 |
+
"loss": 0.0581,
|
| 81 |
+
"step": 120
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.21,
|
| 85 |
+
"learning_rate": 3.571428571428572e-05,
|
| 86 |
+
"loss": 0.0281,
|
| 87 |
+
"step": 130
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.23,
|
| 91 |
+
"learning_rate": 3.846153846153846e-05,
|
| 92 |
+
"loss": 0.0612,
|
| 93 |
+
"step": 140
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.25,
|
| 97 |
+
"learning_rate": 4.120879120879121e-05,
|
| 98 |
+
"loss": 0.0297,
|
| 99 |
+
"step": 150
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.26,
|
| 103 |
+
"learning_rate": 4.3956043956043955e-05,
|
| 104 |
+
"loss": 0.021,
|
| 105 |
+
"step": 160
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.28,
|
| 109 |
+
"learning_rate": 4.670329670329671e-05,
|
| 110 |
+
"loss": 0.019,
|
| 111 |
+
"step": 170
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.3,
|
| 115 |
+
"learning_rate": 4.945054945054945e-05,
|
| 116 |
+
"loss": 0.0233,
|
| 117 |
+
"step": 180
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.31,
|
| 121 |
+
"learning_rate": 4.9755052051439074e-05,
|
| 122 |
+
"loss": 0.0298,
|
| 123 |
+
"step": 190
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.33,
|
| 127 |
+
"learning_rate": 4.944886711573791e-05,
|
| 128 |
+
"loss": 0.0421,
|
| 129 |
+
"step": 200
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.35,
|
| 133 |
+
"learning_rate": 4.914268218003674e-05,
|
| 134 |
+
"loss": 0.0277,
|
| 135 |
+
"step": 210
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.36,
|
| 139 |
+
"learning_rate": 4.883649724433558e-05,
|
| 140 |
+
"loss": 0.0361,
|
| 141 |
+
"step": 220
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.38,
|
| 145 |
+
"learning_rate": 4.8530312308634416e-05,
|
| 146 |
+
"loss": 0.0579,
|
| 147 |
+
"step": 230
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.4,
|
| 151 |
+
"learning_rate": 4.822412737293325e-05,
|
| 152 |
+
"loss": 0.0338,
|
| 153 |
+
"step": 240
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.41,
|
| 157 |
+
"learning_rate": 4.791794243723209e-05,
|
| 158 |
+
"loss": 0.0366,
|
| 159 |
+
"step": 250
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.43,
|
| 163 |
+
"learning_rate": 4.761175750153093e-05,
|
| 164 |
+
"loss": 0.0092,
|
| 165 |
+
"step": 260
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.45,
|
| 169 |
+
"learning_rate": 4.7305572565829766e-05,
|
| 170 |
+
"loss": 0.063,
|
| 171 |
+
"step": 270
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.46,
|
| 175 |
+
"learning_rate": 4.69993876301286e-05,
|
| 176 |
+
"loss": 0.0527,
|
| 177 |
+
"step": 280
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.48,
|
| 181 |
+
"learning_rate": 4.6693202694427433e-05,
|
| 182 |
+
"loss": 0.023,
|
| 183 |
+
"step": 290
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.5,
|
| 187 |
+
"learning_rate": 4.6387017758726274e-05,
|
| 188 |
+
"loss": 0.0437,
|
| 189 |
+
"step": 300
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.51,
|
| 193 |
+
"learning_rate": 4.608083282302511e-05,
|
| 194 |
+
"loss": 0.0381,
|
| 195 |
+
"step": 310
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.53,
|
| 199 |
+
"learning_rate": 4.577464788732395e-05,
|
| 200 |
+
"loss": 0.0288,
|
| 201 |
+
"step": 320
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.55,
|
| 205 |
+
"learning_rate": 4.546846295162278e-05,
|
| 206 |
+
"loss": 0.0385,
|
| 207 |
+
"step": 330
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.56,
|
| 211 |
+
"learning_rate": 4.516227801592162e-05,
|
| 212 |
+
"loss": 0.0442,
|
| 213 |
+
"step": 340
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.58,
|
| 217 |
+
"learning_rate": 4.485609308022046e-05,
|
| 218 |
+
"loss": 0.0128,
|
| 219 |
+
"step": 350
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6,
|
| 223 |
+
"learning_rate": 4.454990814451929e-05,
|
| 224 |
+
"loss": 0.0189,
|
| 225 |
+
"step": 360
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.61,
|
| 229 |
+
"learning_rate": 4.4243723208818125e-05,
|
| 230 |
+
"loss": 0.0219,
|
| 231 |
+
"step": 370
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.63,
|
| 235 |
+
"learning_rate": 4.3937538273116966e-05,
|
| 236 |
+
"loss": 0.0296,
|
| 237 |
+
"step": 380
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.64,
|
| 241 |
+
"learning_rate": 4.363135333741581e-05,
|
| 242 |
+
"loss": 0.058,
|
| 243 |
+
"step": 390
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.66,
|
| 247 |
+
"learning_rate": 4.332516840171464e-05,
|
| 248 |
+
"loss": 0.0317,
|
| 249 |
+
"step": 400
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.68,
|
| 253 |
+
"learning_rate": 4.3018983466013475e-05,
|
| 254 |
+
"loss": 0.0364,
|
| 255 |
+
"step": 410
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.69,
|
| 259 |
+
"learning_rate": 4.271279853031231e-05,
|
| 260 |
+
"loss": 0.0239,
|
| 261 |
+
"step": 420
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.71,
|
| 265 |
+
"learning_rate": 4.240661359461115e-05,
|
| 266 |
+
"loss": 0.0265,
|
| 267 |
+
"step": 430
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.73,
|
| 271 |
+
"learning_rate": 4.210042865890998e-05,
|
| 272 |
+
"loss": 0.0046,
|
| 273 |
+
"step": 440
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.74,
|
| 277 |
+
"learning_rate": 4.179424372320882e-05,
|
| 278 |
+
"loss": 0.0182,
|
| 279 |
+
"step": 450
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.76,
|
| 283 |
+
"learning_rate": 4.148805878750766e-05,
|
| 284 |
+
"loss": 0.0255,
|
| 285 |
+
"step": 460
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.78,
|
| 289 |
+
"learning_rate": 4.118187385180649e-05,
|
| 290 |
+
"loss": 0.0143,
|
| 291 |
+
"step": 470
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.79,
|
| 295 |
+
"learning_rate": 4.087568891610533e-05,
|
| 296 |
+
"loss": 0.0348,
|
| 297 |
+
"step": 480
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.81,
|
| 301 |
+
"learning_rate": 4.0569503980404167e-05,
|
| 302 |
+
"loss": 0.035,
|
| 303 |
+
"step": 490
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.83,
|
| 307 |
+
"learning_rate": 4.0263319044703e-05,
|
| 308 |
+
"loss": 0.0397,
|
| 309 |
+
"step": 500
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.84,
|
| 313 |
+
"learning_rate": 3.9957134109001834e-05,
|
| 314 |
+
"loss": 0.0194,
|
| 315 |
+
"step": 510
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.86,
|
| 319 |
+
"learning_rate": 3.9650949173300675e-05,
|
| 320 |
+
"loss": 0.0075,
|
| 321 |
+
"step": 520
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.88,
|
| 325 |
+
"learning_rate": 3.9344764237599516e-05,
|
| 326 |
+
"loss": 0.0025,
|
| 327 |
+
"step": 530
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.89,
|
| 331 |
+
"learning_rate": 3.903857930189835e-05,
|
| 332 |
+
"loss": 0.002,
|
| 333 |
+
"step": 540
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.91,
|
| 337 |
+
"learning_rate": 3.8732394366197184e-05,
|
| 338 |
+
"loss": 0.0093,
|
| 339 |
+
"step": 550
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.93,
|
| 343 |
+
"learning_rate": 3.8426209430496025e-05,
|
| 344 |
+
"loss": 0.0027,
|
| 345 |
+
"step": 560
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.94,
|
| 349 |
+
"learning_rate": 3.812002449479486e-05,
|
| 350 |
+
"loss": 0.011,
|
| 351 |
+
"step": 570
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 0.96,
|
| 355 |
+
"learning_rate": 3.781383955909369e-05,
|
| 356 |
+
"loss": 0.0133,
|
| 357 |
+
"step": 580
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.98,
|
| 361 |
+
"learning_rate": 3.7507654623392526e-05,
|
| 362 |
+
"loss": 0.0314,
|
| 363 |
+
"step": 590
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.99,
|
| 367 |
+
"learning_rate": 3.720146968769137e-05,
|
| 368 |
+
"loss": 0.0187,
|
| 369 |
+
"step": 600
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.0,
|
| 373 |
+
"eval_accuracy": 0.9973790624531975,
|
| 374 |
+
"eval_loss": 0.010242484509944916,
|
| 375 |
+
"eval_runtime": 251.9013,
|
| 376 |
+
"eval_samples_per_second": 53.013,
|
| 377 |
+
"eval_steps_per_second": 3.315,
|
| 378 |
+
"step": 605
|
| 379 |
+
}
|
| 380 |
+
],
|
| 381 |
+
"logging_steps": 10,
|
| 382 |
+
"max_steps": 1815,
|
| 383 |
+
"num_input_tokens_seen": 0,
|
| 384 |
+
"num_train_epochs": 3,
|
| 385 |
+
"save_steps": 500,
|
| 386 |
+
"total_flos": 3.033148186139259e+18,
|
| 387 |
+
"train_batch_size": 16,
|
| 388 |
+
"trial_name": null,
|
| 389 |
+
"trial_params": null
|
| 390 |
+
}
|