Upload 10 files
Browse files- config.json +30 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +37 -0
- tokenizer_config.json +52 -0
- trainer_state.json +1665 -0
- training_args.bin +3 -0
- vocab.txt +33 -0
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ambrim/attempt8_150M_200k_default_accuracy_mostest",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"EsmForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"emb_layer_norm_before": false,
|
| 9 |
+
"esmfold_config": null,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.0,
|
| 12 |
+
"hidden_size": 640,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 2560,
|
| 15 |
+
"is_folding_model": false,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"mask_token_id": 32,
|
| 18 |
+
"max_position_embeddings": 1026,
|
| 19 |
+
"model_type": "esm",
|
| 20 |
+
"num_attention_heads": 20,
|
| 21 |
+
"num_hidden_layers": 30,
|
| 22 |
+
"pad_token_id": 1,
|
| 23 |
+
"position_embedding_type": "rotary",
|
| 24 |
+
"token_dropout": true,
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.36.0",
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"vocab_list": null,
|
| 29 |
+
"vocab_size": 33
|
| 30 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9f9999f42c5cc66d59d3f8c71fa98466f55c55ad3948cf70cf47cd9faa46b622
|
| 3 |
+
size 593608020
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e4405dd89ac8f6fc978a1d2fc45bb6be16e6a732297900716e2a032df19b582
|
| 3 |
+
size 1182240577
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ee92eb46fa92334cd0695596a0005ba047370328e120afcbe0600b3db0d3cf32
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d011892634a2dfed852b57d4f1ac63def63b8fa9bb88caec3174290a5327251
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "<cls>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<eos>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"mask_token": {
|
| 17 |
+
"content": "<mask>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"pad_token": {
|
| 24 |
+
"content": "<pad>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "<unk>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<cls>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "<eos>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"32": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "<cls>",
|
| 46 |
+
"eos_token": "<eos>",
|
| 47 |
+
"mask_token": "<mask>",
|
| 48 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 49 |
+
"pad_token": "<pad>",
|
| 50 |
+
"tokenizer_class": "EsmTokenizer",
|
| 51 |
+
"unk_token": "<unk>"
|
| 52 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1665 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9558851470160417,
|
| 3 |
+
"best_model_checkpoint": "attempt9_150M_200k_default_accuracy/checkpoint-6862",
|
| 4 |
+
"epoch": 18.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 123516,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 9.963567473039931e-05,
|
| 14 |
+
"loss": 0.0892,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.15,
|
| 19 |
+
"learning_rate": 9.92713494607986e-05,
|
| 20 |
+
"loss": 0.0839,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.22,
|
| 25 |
+
"learning_rate": 9.890702419119791e-05,
|
| 26 |
+
"loss": 0.0938,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.29,
|
| 31 |
+
"learning_rate": 9.854269892159721e-05,
|
| 32 |
+
"loss": 0.0958,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.36,
|
| 37 |
+
"learning_rate": 9.81783736519965e-05,
|
| 38 |
+
"loss": 0.1007,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.44,
|
| 43 |
+
"learning_rate": 9.781404838239581e-05,
|
| 44 |
+
"loss": 0.1044,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.51,
|
| 49 |
+
"learning_rate": 9.744972311279511e-05,
|
| 50 |
+
"loss": 0.1048,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.58,
|
| 55 |
+
"learning_rate": 9.70853978431944e-05,
|
| 56 |
+
"loss": 0.1026,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.66,
|
| 61 |
+
"learning_rate": 9.672107257359371e-05,
|
| 62 |
+
"loss": 0.1084,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.73,
|
| 67 |
+
"learning_rate": 9.635674730399302e-05,
|
| 68 |
+
"loss": 0.1042,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.8,
|
| 73 |
+
"learning_rate": 9.599242203439231e-05,
|
| 74 |
+
"loss": 0.1104,
|
| 75 |
+
"step": 5500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.87,
|
| 79 |
+
"learning_rate": 9.562809676479161e-05,
|
| 80 |
+
"loss": 0.1152,
|
| 81 |
+
"step": 6000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.95,
|
| 85 |
+
"learning_rate": 9.52637714951909e-05,
|
| 86 |
+
"loss": 0.115,
|
| 87 |
+
"step": 6500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.0,
|
| 91 |
+
"eval_accuracy": 0.9558851470160417,
|
| 92 |
+
"eval_loss": 0.11039502918720245,
|
| 93 |
+
"eval_runtime": 141.6598,
|
| 94 |
+
"eval_samples_per_second": 32.296,
|
| 95 |
+
"eval_steps_per_second": 5.386,
|
| 96 |
+
"step": 6862
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.02,
|
| 100 |
+
"learning_rate": 9.489944622559021e-05,
|
| 101 |
+
"loss": 0.1035,
|
| 102 |
+
"step": 7000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.09,
|
| 106 |
+
"learning_rate": 9.45351209559895e-05,
|
| 107 |
+
"loss": 0.0944,
|
| 108 |
+
"step": 7500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.17,
|
| 112 |
+
"learning_rate": 9.417079568638881e-05,
|
| 113 |
+
"loss": 0.0946,
|
| 114 |
+
"step": 8000
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.24,
|
| 118 |
+
"learning_rate": 9.380647041678811e-05,
|
| 119 |
+
"loss": 0.0994,
|
| 120 |
+
"step": 8500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.31,
|
| 124 |
+
"learning_rate": 9.34421451471874e-05,
|
| 125 |
+
"loss": 0.0972,
|
| 126 |
+
"step": 9000
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.38,
|
| 130 |
+
"learning_rate": 9.307781987758671e-05,
|
| 131 |
+
"loss": 0.097,
|
| 132 |
+
"step": 9500
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 1.46,
|
| 136 |
+
"learning_rate": 9.271349460798602e-05,
|
| 137 |
+
"loss": 0.1001,
|
| 138 |
+
"step": 10000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 1.53,
|
| 142 |
+
"learning_rate": 9.234916933838531e-05,
|
| 143 |
+
"loss": 0.1001,
|
| 144 |
+
"step": 10500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.6,
|
| 148 |
+
"learning_rate": 9.198484406878461e-05,
|
| 149 |
+
"loss": 0.0999,
|
| 150 |
+
"step": 11000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.68,
|
| 154 |
+
"learning_rate": 9.162051879918392e-05,
|
| 155 |
+
"loss": 0.1025,
|
| 156 |
+
"step": 11500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.75,
|
| 160 |
+
"learning_rate": 9.125619352958321e-05,
|
| 161 |
+
"loss": 0.1047,
|
| 162 |
+
"step": 12000
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.82,
|
| 166 |
+
"learning_rate": 9.089186825998251e-05,
|
| 167 |
+
"loss": 0.1033,
|
| 168 |
+
"step": 12500
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.89,
|
| 172 |
+
"learning_rate": 9.052754299038182e-05,
|
| 173 |
+
"loss": 0.1115,
|
| 174 |
+
"step": 13000
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.97,
|
| 178 |
+
"learning_rate": 9.016321772078111e-05,
|
| 179 |
+
"loss": 0.1106,
|
| 180 |
+
"step": 13500
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 2.0,
|
| 184 |
+
"eval_accuracy": 0.9502861541550922,
|
| 185 |
+
"eval_loss": 0.13403260707855225,
|
| 186 |
+
"eval_runtime": 141.6928,
|
| 187 |
+
"eval_samples_per_second": 32.288,
|
| 188 |
+
"eval_steps_per_second": 5.385,
|
| 189 |
+
"step": 13724
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.04,
|
| 193 |
+
"learning_rate": 8.979889245118042e-05,
|
| 194 |
+
"loss": 0.0919,
|
| 195 |
+
"step": 14000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.11,
|
| 199 |
+
"learning_rate": 8.943456718157972e-05,
|
| 200 |
+
"loss": 0.0875,
|
| 201 |
+
"step": 14500
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.19,
|
| 205 |
+
"learning_rate": 8.907024191197903e-05,
|
| 206 |
+
"loss": 0.0888,
|
| 207 |
+
"step": 15000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.26,
|
| 211 |
+
"learning_rate": 8.870591664237832e-05,
|
| 212 |
+
"loss": 0.0872,
|
| 213 |
+
"step": 15500
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.33,
|
| 217 |
+
"learning_rate": 8.834159137277762e-05,
|
| 218 |
+
"loss": 0.0966,
|
| 219 |
+
"step": 16000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.4,
|
| 223 |
+
"learning_rate": 8.797726610317693e-05,
|
| 224 |
+
"loss": 0.0985,
|
| 225 |
+
"step": 16500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.48,
|
| 229 |
+
"learning_rate": 8.761294083357622e-05,
|
| 230 |
+
"loss": 0.09,
|
| 231 |
+
"step": 17000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.55,
|
| 235 |
+
"learning_rate": 8.724861556397553e-05,
|
| 236 |
+
"loss": 0.0896,
|
| 237 |
+
"step": 17500
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.62,
|
| 241 |
+
"learning_rate": 8.688429029437483e-05,
|
| 242 |
+
"loss": 0.0995,
|
| 243 |
+
"step": 18000
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.7,
|
| 247 |
+
"learning_rate": 8.651996502477412e-05,
|
| 248 |
+
"loss": 0.099,
|
| 249 |
+
"step": 18500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.77,
|
| 253 |
+
"learning_rate": 8.615563975517343e-05,
|
| 254 |
+
"loss": 0.0973,
|
| 255 |
+
"step": 19000
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.84,
|
| 259 |
+
"learning_rate": 8.579131448557272e-05,
|
| 260 |
+
"loss": 0.1039,
|
| 261 |
+
"step": 19500
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.91,
|
| 265 |
+
"learning_rate": 8.542698921597203e-05,
|
| 266 |
+
"loss": 0.0977,
|
| 267 |
+
"step": 20000
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.99,
|
| 271 |
+
"learning_rate": 8.506266394637132e-05,
|
| 272 |
+
"loss": 0.096,
|
| 273 |
+
"step": 20500
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 3.0,
|
| 277 |
+
"eval_accuracy": 0.9471576383722038,
|
| 278 |
+
"eval_loss": 0.15382222831249237,
|
| 279 |
+
"eval_runtime": 141.7395,
|
| 280 |
+
"eval_samples_per_second": 32.278,
|
| 281 |
+
"eval_steps_per_second": 5.383,
|
| 282 |
+
"step": 20586
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 3.06,
|
| 286 |
+
"learning_rate": 8.469833867677062e-05,
|
| 287 |
+
"loss": 0.0867,
|
| 288 |
+
"step": 21000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.13,
|
| 292 |
+
"learning_rate": 8.433401340716991e-05,
|
| 293 |
+
"loss": 0.0764,
|
| 294 |
+
"step": 21500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 3.21,
|
| 298 |
+
"learning_rate": 8.396968813756922e-05,
|
| 299 |
+
"loss": 0.0891,
|
| 300 |
+
"step": 22000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 3.28,
|
| 304 |
+
"learning_rate": 8.360536286796853e-05,
|
| 305 |
+
"loss": 0.0837,
|
| 306 |
+
"step": 22500
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 3.35,
|
| 310 |
+
"learning_rate": 8.324103759836782e-05,
|
| 311 |
+
"loss": 0.0865,
|
| 312 |
+
"step": 23000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 3.42,
|
| 316 |
+
"learning_rate": 8.287671232876712e-05,
|
| 317 |
+
"loss": 0.0895,
|
| 318 |
+
"step": 23500
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 3.5,
|
| 322 |
+
"learning_rate": 8.251238705916643e-05,
|
| 323 |
+
"loss": 0.0883,
|
| 324 |
+
"step": 24000
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 3.57,
|
| 328 |
+
"learning_rate": 8.214806178956573e-05,
|
| 329 |
+
"loss": 0.0896,
|
| 330 |
+
"step": 24500
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.64,
|
| 334 |
+
"learning_rate": 8.178373651996502e-05,
|
| 335 |
+
"loss": 0.0896,
|
| 336 |
+
"step": 25000
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 3.72,
|
| 340 |
+
"learning_rate": 8.141941125036433e-05,
|
| 341 |
+
"loss": 0.0939,
|
| 342 |
+
"step": 25500
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 3.79,
|
| 346 |
+
"learning_rate": 8.105508598076364e-05,
|
| 347 |
+
"loss": 0.0906,
|
| 348 |
+
"step": 26000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 3.86,
|
| 352 |
+
"learning_rate": 8.069076071116293e-05,
|
| 353 |
+
"loss": 0.0927,
|
| 354 |
+
"step": 26500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 3.93,
|
| 358 |
+
"learning_rate": 8.032643544156223e-05,
|
| 359 |
+
"loss": 0.0982,
|
| 360 |
+
"step": 27000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 4.0,
|
| 364 |
+
"eval_accuracy": 0.9438823768687553,
|
| 365 |
+
"eval_loss": 0.1686604619026184,
|
| 366 |
+
"eval_runtime": 141.8449,
|
| 367 |
+
"eval_samples_per_second": 32.254,
|
| 368 |
+
"eval_steps_per_second": 5.379,
|
| 369 |
+
"step": 27448
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 4.01,
|
| 373 |
+
"learning_rate": 7.996211017196154e-05,
|
| 374 |
+
"loss": 0.0903,
|
| 375 |
+
"step": 27500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 4.08,
|
| 379 |
+
"learning_rate": 7.959778490236083e-05,
|
| 380 |
+
"loss": 0.0753,
|
| 381 |
+
"step": 28000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 4.15,
|
| 385 |
+
"learning_rate": 7.923345963276013e-05,
|
| 386 |
+
"loss": 0.0783,
|
| 387 |
+
"step": 28500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 4.23,
|
| 391 |
+
"learning_rate": 7.886913436315944e-05,
|
| 392 |
+
"loss": 0.0776,
|
| 393 |
+
"step": 29000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 4.3,
|
| 397 |
+
"learning_rate": 7.850480909355873e-05,
|
| 398 |
+
"loss": 0.0826,
|
| 399 |
+
"step": 29500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.37,
|
| 403 |
+
"learning_rate": 7.814048382395804e-05,
|
| 404 |
+
"loss": 0.0771,
|
| 405 |
+
"step": 30000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 4.44,
|
| 409 |
+
"learning_rate": 7.777615855435734e-05,
|
| 410 |
+
"loss": 0.0879,
|
| 411 |
+
"step": 30500
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 4.52,
|
| 415 |
+
"learning_rate": 7.741183328475663e-05,
|
| 416 |
+
"loss": 0.0877,
|
| 417 |
+
"step": 31000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.59,
|
| 421 |
+
"learning_rate": 7.704750801515594e-05,
|
| 422 |
+
"loss": 0.0842,
|
| 423 |
+
"step": 31500
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.66,
|
| 427 |
+
"learning_rate": 7.668318274555523e-05,
|
| 428 |
+
"loss": 0.088,
|
| 429 |
+
"step": 32000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 4.74,
|
| 433 |
+
"learning_rate": 7.631885747595454e-05,
|
| 434 |
+
"loss": 0.0805,
|
| 435 |
+
"step": 32500
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.81,
|
| 439 |
+
"learning_rate": 7.595453220635383e-05,
|
| 440 |
+
"loss": 0.0863,
|
| 441 |
+
"step": 33000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.88,
|
| 445 |
+
"learning_rate": 7.559020693675313e-05,
|
| 446 |
+
"loss": 0.0895,
|
| 447 |
+
"step": 33500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 4.95,
|
| 451 |
+
"learning_rate": 7.522588166715244e-05,
|
| 452 |
+
"loss": 0.092,
|
| 453 |
+
"step": 34000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 5.0,
|
| 457 |
+
"eval_accuracy": 0.9422295566477099,
|
| 458 |
+
"eval_loss": 0.1879999339580536,
|
| 459 |
+
"eval_runtime": 141.6558,
|
| 460 |
+
"eval_samples_per_second": 32.297,
|
| 461 |
+
"eval_steps_per_second": 5.386,
|
| 462 |
+
"step": 34310
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 5.03,
|
| 466 |
+
"learning_rate": 7.486155639755173e-05,
|
| 467 |
+
"loss": 0.0807,
|
| 468 |
+
"step": 34500
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 5.1,
|
| 472 |
+
"learning_rate": 7.449723112795104e-05,
|
| 473 |
+
"loss": 0.0698,
|
| 474 |
+
"step": 35000
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 5.17,
|
| 478 |
+
"learning_rate": 7.413290585835034e-05,
|
| 479 |
+
"loss": 0.0766,
|
| 480 |
+
"step": 35500
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 5.25,
|
| 484 |
+
"learning_rate": 7.376858058874963e-05,
|
| 485 |
+
"loss": 0.0733,
|
| 486 |
+
"step": 36000
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 5.32,
|
| 490 |
+
"learning_rate": 7.340425531914894e-05,
|
| 491 |
+
"loss": 0.0727,
|
| 492 |
+
"step": 36500
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 5.39,
|
| 496 |
+
"learning_rate": 7.303993004954824e-05,
|
| 497 |
+
"loss": 0.0762,
|
| 498 |
+
"step": 37000
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 5.46,
|
| 502 |
+
"learning_rate": 7.267560477994753e-05,
|
| 503 |
+
"loss": 0.0764,
|
| 504 |
+
"step": 37500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 5.54,
|
| 508 |
+
"learning_rate": 7.231127951034684e-05,
|
| 509 |
+
"loss": 0.0833,
|
| 510 |
+
"step": 38000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 5.61,
|
| 514 |
+
"learning_rate": 7.194695424074615e-05,
|
| 515 |
+
"loss": 0.0792,
|
| 516 |
+
"step": 38500
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 5.68,
|
| 520 |
+
"learning_rate": 7.158262897114544e-05,
|
| 521 |
+
"loss": 0.0816,
|
| 522 |
+
"step": 39000
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 5.76,
|
| 526 |
+
"learning_rate": 7.121830370154474e-05,
|
| 527 |
+
"loss": 0.082,
|
| 528 |
+
"step": 39500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 5.83,
|
| 532 |
+
"learning_rate": 7.085397843194405e-05,
|
| 533 |
+
"loss": 0.08,
|
| 534 |
+
"step": 40000
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 5.9,
|
| 538 |
+
"learning_rate": 7.048965316234334e-05,
|
| 539 |
+
"loss": 0.0827,
|
| 540 |
+
"step": 40500
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 5.97,
|
| 544 |
+
"learning_rate": 7.012532789274264e-05,
|
| 545 |
+
"loss": 0.0808,
|
| 546 |
+
"step": 41000
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 6.0,
|
| 550 |
+
"eval_accuracy": 0.9389187672329679,
|
| 551 |
+
"eval_loss": 0.22079767286777496,
|
| 552 |
+
"eval_runtime": 141.675,
|
| 553 |
+
"eval_samples_per_second": 32.292,
|
| 554 |
+
"eval_steps_per_second": 5.386,
|
| 555 |
+
"step": 41172
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 6.05,
|
| 559 |
+
"learning_rate": 6.976100262314195e-05,
|
| 560 |
+
"loss": 0.076,
|
| 561 |
+
"step": 41500
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 6.12,
|
| 565 |
+
"learning_rate": 6.939667735354124e-05,
|
| 566 |
+
"loss": 0.0736,
|
| 567 |
+
"step": 42000
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 6.19,
|
| 571 |
+
"learning_rate": 6.903235208394055e-05,
|
| 572 |
+
"loss": 0.0717,
|
| 573 |
+
"step": 42500
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 6.27,
|
| 577 |
+
"learning_rate": 6.866802681433985e-05,
|
| 578 |
+
"loss": 0.0711,
|
| 579 |
+
"step": 43000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 6.34,
|
| 583 |
+
"learning_rate": 6.830370154473916e-05,
|
| 584 |
+
"loss": 0.0715,
|
| 585 |
+
"step": 43500
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 6.41,
|
| 589 |
+
"learning_rate": 6.793937627513845e-05,
|
| 590 |
+
"loss": 0.0713,
|
| 591 |
+
"step": 44000
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 6.48,
|
| 595 |
+
"learning_rate": 6.757505100553775e-05,
|
| 596 |
+
"loss": 0.0728,
|
| 597 |
+
"step": 44500
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 6.56,
|
| 601 |
+
"learning_rate": 6.721072573593705e-05,
|
| 602 |
+
"loss": 0.0735,
|
| 603 |
+
"step": 45000
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 6.63,
|
| 607 |
+
"learning_rate": 6.684640046633635e-05,
|
| 608 |
+
"loss": 0.0789,
|
| 609 |
+
"step": 45500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 6.7,
|
| 613 |
+
"learning_rate": 6.648207519673564e-05,
|
| 614 |
+
"loss": 0.0733,
|
| 615 |
+
"step": 46000
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 6.78,
|
| 619 |
+
"learning_rate": 6.611774992713495e-05,
|
| 620 |
+
"loss": 0.0733,
|
| 621 |
+
"step": 46500
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 6.85,
|
| 625 |
+
"learning_rate": 6.575342465753424e-05,
|
| 626 |
+
"loss": 0.0786,
|
| 627 |
+
"step": 47000
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 6.92,
|
| 631 |
+
"learning_rate": 6.538909938793355e-05,
|
| 632 |
+
"loss": 0.0752,
|
| 633 |
+
"step": 47500
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 7.0,
|
| 637 |
+
"learning_rate": 6.502477411833285e-05,
|
| 638 |
+
"loss": 0.0818,
|
| 639 |
+
"step": 48000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 7.0,
|
| 643 |
+
"eval_accuracy": 0.940737899270648,
|
| 644 |
+
"eval_loss": 0.23214486241340637,
|
| 645 |
+
"eval_runtime": 141.6638,
|
| 646 |
+
"eval_samples_per_second": 32.295,
|
| 647 |
+
"eval_steps_per_second": 5.386,
|
| 648 |
+
"step": 48034
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 7.07,
|
| 652 |
+
"learning_rate": 6.466044884873214e-05,
|
| 653 |
+
"loss": 0.0642,
|
| 654 |
+
"step": 48500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 7.14,
|
| 658 |
+
"learning_rate": 6.429612357913145e-05,
|
| 659 |
+
"loss": 0.0668,
|
| 660 |
+
"step": 49000
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 7.21,
|
| 664 |
+
"learning_rate": 6.393179830953075e-05,
|
| 665 |
+
"loss": 0.0648,
|
| 666 |
+
"step": 49500
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 7.29,
|
| 670 |
+
"learning_rate": 6.356747303993004e-05,
|
| 671 |
+
"loss": 0.0689,
|
| 672 |
+
"step": 50000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 7.36,
|
| 676 |
+
"learning_rate": 6.320314777032935e-05,
|
| 677 |
+
"loss": 0.0648,
|
| 678 |
+
"step": 50500
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 7.43,
|
| 682 |
+
"learning_rate": 6.283882250072866e-05,
|
| 683 |
+
"loss": 0.0697,
|
| 684 |
+
"step": 51000
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 7.51,
|
| 688 |
+
"learning_rate": 6.247449723112795e-05,
|
| 689 |
+
"loss": 0.0685,
|
| 690 |
+
"step": 51500
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 7.58,
|
| 694 |
+
"learning_rate": 6.211017196152725e-05,
|
| 695 |
+
"loss": 0.0719,
|
| 696 |
+
"step": 52000
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 7.65,
|
| 700 |
+
"learning_rate": 6.174584669192656e-05,
|
| 701 |
+
"loss": 0.0698,
|
| 702 |
+
"step": 52500
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 7.72,
|
| 706 |
+
"learning_rate": 6.138152142232586e-05,
|
| 707 |
+
"loss": 0.0744,
|
| 708 |
+
"step": 53000
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 7.8,
|
| 712 |
+
"learning_rate": 6.1017196152725155e-05,
|
| 713 |
+
"loss": 0.0753,
|
| 714 |
+
"step": 53500
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 7.87,
|
| 718 |
+
"learning_rate": 6.065287088312446e-05,
|
| 719 |
+
"loss": 0.0708,
|
| 720 |
+
"step": 54000
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 7.94,
|
| 724 |
+
"learning_rate": 6.028854561352376e-05,
|
| 725 |
+
"loss": 0.0741,
|
| 726 |
+
"step": 54500
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 8.0,
|
| 730 |
+
"eval_accuracy": 0.9376047494123735,
|
| 731 |
+
"eval_loss": 0.2665192186832428,
|
| 732 |
+
"eval_runtime": 141.9131,
|
| 733 |
+
"eval_samples_per_second": 32.238,
|
| 734 |
+
"eval_steps_per_second": 5.377,
|
| 735 |
+
"step": 54896
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 8.02,
|
| 739 |
+
"learning_rate": 5.992422034392306e-05,
|
| 740 |
+
"loss": 0.0685,
|
| 741 |
+
"step": 55000
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 8.09,
|
| 745 |
+
"learning_rate": 5.9559895074322355e-05,
|
| 746 |
+
"loss": 0.0597,
|
| 747 |
+
"step": 55500
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 8.16,
|
| 751 |
+
"learning_rate": 5.919556980472166e-05,
|
| 752 |
+
"loss": 0.0653,
|
| 753 |
+
"step": 56000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 8.23,
|
| 757 |
+
"learning_rate": 5.883124453512095e-05,
|
| 758 |
+
"loss": 0.0713,
|
| 759 |
+
"step": 56500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 8.31,
|
| 763 |
+
"learning_rate": 5.846691926552026e-05,
|
| 764 |
+
"loss": 0.0648,
|
| 765 |
+
"step": 57000
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 8.38,
|
| 769 |
+
"learning_rate": 5.810259399591956e-05,
|
| 770 |
+
"loss": 0.0676,
|
| 771 |
+
"step": 57500
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 8.45,
|
| 775 |
+
"learning_rate": 5.7738268726318855e-05,
|
| 776 |
+
"loss": 0.0608,
|
| 777 |
+
"step": 58000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 8.53,
|
| 781 |
+
"learning_rate": 5.737394345671816e-05,
|
| 782 |
+
"loss": 0.0711,
|
| 783 |
+
"step": 58500
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 8.6,
|
| 787 |
+
"learning_rate": 5.7009618187117465e-05,
|
| 788 |
+
"loss": 0.0661,
|
| 789 |
+
"step": 59000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 8.67,
|
| 793 |
+
"learning_rate": 5.664529291751676e-05,
|
| 794 |
+
"loss": 0.0691,
|
| 795 |
+
"step": 59500
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 8.74,
|
| 799 |
+
"learning_rate": 5.628096764791606e-05,
|
| 800 |
+
"loss": 0.0717,
|
| 801 |
+
"step": 60000
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 8.82,
|
| 805 |
+
"learning_rate": 5.591664237831537e-05,
|
| 806 |
+
"loss": 0.0652,
|
| 807 |
+
"step": 60500
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 8.89,
|
| 811 |
+
"learning_rate": 5.555231710871466e-05,
|
| 812 |
+
"loss": 0.0641,
|
| 813 |
+
"step": 61000
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 8.96,
|
| 817 |
+
"learning_rate": 5.5187991839113965e-05,
|
| 818 |
+
"loss": 0.0681,
|
| 819 |
+
"step": 61500
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 9.0,
|
| 823 |
+
"eval_accuracy": 0.936189811727815,
|
| 824 |
+
"eval_loss": 0.24231955409049988,
|
| 825 |
+
"eval_runtime": 141.9857,
|
| 826 |
+
"eval_samples_per_second": 32.222,
|
| 827 |
+
"eval_steps_per_second": 5.374,
|
| 828 |
+
"step": 61758
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 9.04,
|
| 832 |
+
"learning_rate": 5.482366656951326e-05,
|
| 833 |
+
"loss": 0.0606,
|
| 834 |
+
"step": 62000
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 9.11,
|
| 838 |
+
"learning_rate": 5.445934129991257e-05,
|
| 839 |
+
"loss": 0.0574,
|
| 840 |
+
"step": 62500
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 9.18,
|
| 844 |
+
"learning_rate": 5.409501603031186e-05,
|
| 845 |
+
"loss": 0.0595,
|
| 846 |
+
"step": 63000
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 9.25,
|
| 850 |
+
"learning_rate": 5.3730690760711165e-05,
|
| 851 |
+
"loss": 0.0604,
|
| 852 |
+
"step": 63500
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 9.33,
|
| 856 |
+
"learning_rate": 5.336636549111047e-05,
|
| 857 |
+
"loss": 0.0592,
|
| 858 |
+
"step": 64000
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 9.4,
|
| 862 |
+
"learning_rate": 5.300204022150976e-05,
|
| 863 |
+
"loss": 0.0614,
|
| 864 |
+
"step": 64500
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 9.47,
|
| 868 |
+
"learning_rate": 5.263771495190907e-05,
|
| 869 |
+
"loss": 0.066,
|
| 870 |
+
"step": 65000
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 9.55,
|
| 874 |
+
"learning_rate": 5.227338968230837e-05,
|
| 875 |
+
"loss": 0.0671,
|
| 876 |
+
"step": 65500
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 9.62,
|
| 880 |
+
"learning_rate": 5.1909064412707665e-05,
|
| 881 |
+
"loss": 0.0671,
|
| 882 |
+
"step": 66000
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 9.69,
|
| 886 |
+
"learning_rate": 5.154473914310697e-05,
|
| 887 |
+
"loss": 0.064,
|
| 888 |
+
"step": 66500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 9.76,
|
| 892 |
+
"learning_rate": 5.1180413873506275e-05,
|
| 893 |
+
"loss": 0.0675,
|
| 894 |
+
"step": 67000
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 9.84,
|
| 898 |
+
"learning_rate": 5.081608860390557e-05,
|
| 899 |
+
"loss": 0.062,
|
| 900 |
+
"step": 67500
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 9.91,
|
| 904 |
+
"learning_rate": 5.045176333430487e-05,
|
| 905 |
+
"loss": 0.0669,
|
| 906 |
+
"step": 68000
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 9.98,
|
| 910 |
+
"learning_rate": 5.008743806470417e-05,
|
| 911 |
+
"loss": 0.0644,
|
| 912 |
+
"step": 68500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 10.0,
|
| 916 |
+
"eval_accuracy": 0.934086971297052,
|
| 917 |
+
"eval_loss": 0.2860284447669983,
|
| 918 |
+
"eval_runtime": 142.0094,
|
| 919 |
+
"eval_samples_per_second": 32.216,
|
| 920 |
+
"eval_steps_per_second": 5.373,
|
| 921 |
+
"step": 68620
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 10.06,
|
| 925 |
+
"learning_rate": 4.972311279510347e-05,
|
| 926 |
+
"loss": 0.0594,
|
| 927 |
+
"step": 69000
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 10.13,
|
| 931 |
+
"learning_rate": 4.935878752550277e-05,
|
| 932 |
+
"loss": 0.056,
|
| 933 |
+
"step": 69500
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 10.2,
|
| 937 |
+
"learning_rate": 4.899446225590207e-05,
|
| 938 |
+
"loss": 0.0549,
|
| 939 |
+
"step": 70000
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 10.27,
|
| 943 |
+
"learning_rate": 4.863013698630137e-05,
|
| 944 |
+
"loss": 0.0566,
|
| 945 |
+
"step": 70500
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 10.35,
|
| 949 |
+
"learning_rate": 4.826581171670067e-05,
|
| 950 |
+
"loss": 0.0587,
|
| 951 |
+
"step": 71000
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 10.42,
|
| 955 |
+
"learning_rate": 4.7901486447099975e-05,
|
| 956 |
+
"loss": 0.0585,
|
| 957 |
+
"step": 71500
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 10.49,
|
| 961 |
+
"learning_rate": 4.7537161177499274e-05,
|
| 962 |
+
"loss": 0.0627,
|
| 963 |
+
"step": 72000
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 10.57,
|
| 967 |
+
"learning_rate": 4.717283590789857e-05,
|
| 968 |
+
"loss": 0.0557,
|
| 969 |
+
"step": 72500
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 10.64,
|
| 973 |
+
"learning_rate": 4.680851063829788e-05,
|
| 974 |
+
"loss": 0.058,
|
| 975 |
+
"step": 73000
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 10.71,
|
| 979 |
+
"learning_rate": 4.6444185368697176e-05,
|
| 980 |
+
"loss": 0.0616,
|
| 981 |
+
"step": 73500
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 10.78,
|
| 985 |
+
"learning_rate": 4.6079860099096475e-05,
|
| 986 |
+
"loss": 0.0637,
|
| 987 |
+
"step": 74000
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 10.86,
|
| 991 |
+
"learning_rate": 4.571553482949577e-05,
|
| 992 |
+
"loss": 0.0646,
|
| 993 |
+
"step": 74500
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 10.93,
|
| 997 |
+
"learning_rate": 4.535120955989507e-05,
|
| 998 |
+
"loss": 0.0598,
|
| 999 |
+
"step": 75000
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 11.0,
|
| 1003 |
+
"eval_accuracy": 0.9335947295115942,
|
| 1004 |
+
"eval_loss": 0.30142533779144287,
|
| 1005 |
+
"eval_runtime": 141.9752,
|
| 1006 |
+
"eval_samples_per_second": 32.224,
|
| 1007 |
+
"eval_steps_per_second": 5.374,
|
| 1008 |
+
"step": 75482
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 11.0,
|
| 1012 |
+
"learning_rate": 4.498688429029438e-05,
|
| 1013 |
+
"loss": 0.063,
|
| 1014 |
+
"step": 75500
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 11.08,
|
| 1018 |
+
"learning_rate": 4.4622559020693675e-05,
|
| 1019 |
+
"loss": 0.0552,
|
| 1020 |
+
"step": 76000
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 11.15,
|
| 1024 |
+
"learning_rate": 4.4258233751092974e-05,
|
| 1025 |
+
"loss": 0.0521,
|
| 1026 |
+
"step": 76500
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 11.22,
|
| 1030 |
+
"learning_rate": 4.389390848149228e-05,
|
| 1031 |
+
"loss": 0.0526,
|
| 1032 |
+
"step": 77000
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 11.29,
|
| 1036 |
+
"learning_rate": 4.352958321189158e-05,
|
| 1037 |
+
"loss": 0.0575,
|
| 1038 |
+
"step": 77500
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 11.37,
|
| 1042 |
+
"learning_rate": 4.316525794229088e-05,
|
| 1043 |
+
"loss": 0.0564,
|
| 1044 |
+
"step": 78000
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 11.44,
|
| 1048 |
+
"learning_rate": 4.280093267269018e-05,
|
| 1049 |
+
"loss": 0.0574,
|
| 1050 |
+
"step": 78500
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 11.51,
|
| 1054 |
+
"learning_rate": 4.243660740308948e-05,
|
| 1055 |
+
"loss": 0.0609,
|
| 1056 |
+
"step": 79000
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 11.59,
|
| 1060 |
+
"learning_rate": 4.2072282133488785e-05,
|
| 1061 |
+
"loss": 0.0524,
|
| 1062 |
+
"step": 79500
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 11.66,
|
| 1066 |
+
"learning_rate": 4.1707956863888084e-05,
|
| 1067 |
+
"loss": 0.0552,
|
| 1068 |
+
"step": 80000
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 11.73,
|
| 1072 |
+
"learning_rate": 4.134363159428738e-05,
|
| 1073 |
+
"loss": 0.0596,
|
| 1074 |
+
"step": 80500
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 11.8,
|
| 1078 |
+
"learning_rate": 4.097930632468668e-05,
|
| 1079 |
+
"loss": 0.0535,
|
| 1080 |
+
"step": 81000
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 11.88,
|
| 1084 |
+
"learning_rate": 4.061498105508598e-05,
|
| 1085 |
+
"loss": 0.0672,
|
| 1086 |
+
"step": 81500
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 11.95,
|
| 1090 |
+
"learning_rate": 4.025065578548528e-05,
|
| 1091 |
+
"loss": 0.0609,
|
| 1092 |
+
"step": 82000
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 12.0,
|
| 1096 |
+
"eval_accuracy": 0.9334938096476301,
|
| 1097 |
+
"eval_loss": 0.33718985319137573,
|
| 1098 |
+
"eval_runtime": 142.0149,
|
| 1099 |
+
"eval_samples_per_second": 32.215,
|
| 1100 |
+
"eval_steps_per_second": 5.373,
|
| 1101 |
+
"step": 82344
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 12.02,
|
| 1105 |
+
"learning_rate": 3.988633051588458e-05,
|
| 1106 |
+
"loss": 0.0527,
|
| 1107 |
+
"step": 82500
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 12.1,
|
| 1111 |
+
"learning_rate": 3.952200524628388e-05,
|
| 1112 |
+
"loss": 0.051,
|
| 1113 |
+
"step": 83000
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 12.17,
|
| 1117 |
+
"learning_rate": 3.915767997668319e-05,
|
| 1118 |
+
"loss": 0.0544,
|
| 1119 |
+
"step": 83500
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 12.24,
|
| 1123 |
+
"learning_rate": 3.8793354707082485e-05,
|
| 1124 |
+
"loss": 0.0507,
|
| 1125 |
+
"step": 84000
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 12.31,
|
| 1129 |
+
"learning_rate": 3.8429029437481784e-05,
|
| 1130 |
+
"loss": 0.0563,
|
| 1131 |
+
"step": 84500
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 12.39,
|
| 1135 |
+
"learning_rate": 3.806470416788109e-05,
|
| 1136 |
+
"loss": 0.0499,
|
| 1137 |
+
"step": 85000
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 12.46,
|
| 1141 |
+
"learning_rate": 3.770037889828039e-05,
|
| 1142 |
+
"loss": 0.0621,
|
| 1143 |
+
"step": 85500
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 12.53,
|
| 1147 |
+
"learning_rate": 3.7336053628679686e-05,
|
| 1148 |
+
"loss": 0.0579,
|
| 1149 |
+
"step": 86000
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 12.61,
|
| 1153 |
+
"learning_rate": 3.697172835907899e-05,
|
| 1154 |
+
"loss": 0.0507,
|
| 1155 |
+
"step": 86500
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 12.68,
|
| 1159 |
+
"learning_rate": 3.660740308947829e-05,
|
| 1160 |
+
"loss": 0.054,
|
| 1161 |
+
"step": 87000
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 12.75,
|
| 1165 |
+
"learning_rate": 3.624307781987759e-05,
|
| 1166 |
+
"loss": 0.0538,
|
| 1167 |
+
"step": 87500
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 12.82,
|
| 1171 |
+
"learning_rate": 3.587875255027689e-05,
|
| 1172 |
+
"loss": 0.0512,
|
| 1173 |
+
"step": 88000
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 12.9,
|
| 1177 |
+
"learning_rate": 3.5514427280676185e-05,
|
| 1178 |
+
"loss": 0.0548,
|
| 1179 |
+
"step": 88500
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 12.97,
|
| 1183 |
+
"learning_rate": 3.515010201107549e-05,
|
| 1184 |
+
"loss": 0.0525,
|
| 1185 |
+
"step": 89000
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 13.0,
|
| 1189 |
+
"eval_accuracy": 0.9331416199182858,
|
| 1190 |
+
"eval_loss": 0.3078489601612091,
|
| 1191 |
+
"eval_runtime": 141.952,
|
| 1192 |
+
"eval_samples_per_second": 32.229,
|
| 1193 |
+
"eval_steps_per_second": 5.375,
|
| 1194 |
+
"step": 89206
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 13.04,
|
| 1198 |
+
"learning_rate": 3.478577674147479e-05,
|
| 1199 |
+
"loss": 0.054,
|
| 1200 |
+
"step": 89500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 13.12,
|
| 1204 |
+
"learning_rate": 3.442145147187409e-05,
|
| 1205 |
+
"loss": 0.0486,
|
| 1206 |
+
"step": 90000
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 13.19,
|
| 1210 |
+
"learning_rate": 3.405712620227339e-05,
|
| 1211 |
+
"loss": 0.0524,
|
| 1212 |
+
"step": 90500
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 13.26,
|
| 1216 |
+
"learning_rate": 3.369280093267269e-05,
|
| 1217 |
+
"loss": 0.0503,
|
| 1218 |
+
"step": 91000
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 13.33,
|
| 1222 |
+
"learning_rate": 3.332847566307199e-05,
|
| 1223 |
+
"loss": 0.0494,
|
| 1224 |
+
"step": 91500
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 13.41,
|
| 1228 |
+
"learning_rate": 3.2964150393471295e-05,
|
| 1229 |
+
"loss": 0.0465,
|
| 1230 |
+
"step": 92000
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 13.48,
|
| 1234 |
+
"learning_rate": 3.2599825123870594e-05,
|
| 1235 |
+
"loss": 0.0503,
|
| 1236 |
+
"step": 92500
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 13.55,
|
| 1240 |
+
"learning_rate": 3.22354998542699e-05,
|
| 1241 |
+
"loss": 0.0554,
|
| 1242 |
+
"step": 93000
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 13.63,
|
| 1246 |
+
"learning_rate": 3.18711745846692e-05,
|
| 1247 |
+
"loss": 0.0497,
|
| 1248 |
+
"step": 93500
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 13.7,
|
| 1252 |
+
"learning_rate": 3.1506849315068496e-05,
|
| 1253 |
+
"loss": 0.0509,
|
| 1254 |
+
"step": 94000
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 13.77,
|
| 1258 |
+
"learning_rate": 3.1142524045467795e-05,
|
| 1259 |
+
"loss": 0.0566,
|
| 1260 |
+
"step": 94500
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 13.84,
|
| 1264 |
+
"learning_rate": 3.077819877586709e-05,
|
| 1265 |
+
"loss": 0.0535,
|
| 1266 |
+
"step": 95000
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 13.92,
|
| 1270 |
+
"learning_rate": 3.0413873506266395e-05,
|
| 1271 |
+
"loss": 0.0545,
|
| 1272 |
+
"step": 95500
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 13.99,
|
| 1276 |
+
"learning_rate": 3.0049548236665697e-05,
|
| 1277 |
+
"loss": 0.049,
|
| 1278 |
+
"step": 96000
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 14.0,
|
| 1282 |
+
"eval_accuracy": 0.9329361759095016,
|
| 1283 |
+
"eval_loss": 0.35664141178131104,
|
| 1284 |
+
"eval_runtime": 141.9778,
|
| 1285 |
+
"eval_samples_per_second": 32.223,
|
| 1286 |
+
"eval_steps_per_second": 5.374,
|
| 1287 |
+
"step": 96068
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 14.06,
|
| 1291 |
+
"learning_rate": 2.9685222967064995e-05,
|
| 1292 |
+
"loss": 0.0496,
|
| 1293 |
+
"step": 96500
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 14.14,
|
| 1297 |
+
"learning_rate": 2.93208976974643e-05,
|
| 1298 |
+
"loss": 0.0493,
|
| 1299 |
+
"step": 97000
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 14.21,
|
| 1303 |
+
"learning_rate": 2.89565724278636e-05,
|
| 1304 |
+
"loss": 0.0472,
|
| 1305 |
+
"step": 97500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 14.28,
|
| 1309 |
+
"learning_rate": 2.8592247158262898e-05,
|
| 1310 |
+
"loss": 0.0499,
|
| 1311 |
+
"step": 98000
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 14.35,
|
| 1315 |
+
"learning_rate": 2.82279218886622e-05,
|
| 1316 |
+
"loss": 0.0481,
|
| 1317 |
+
"step": 98500
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 14.43,
|
| 1321 |
+
"learning_rate": 2.7863596619061498e-05,
|
| 1322 |
+
"loss": 0.0495,
|
| 1323 |
+
"step": 99000
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 14.5,
|
| 1327 |
+
"learning_rate": 2.7499271349460797e-05,
|
| 1328 |
+
"loss": 0.049,
|
| 1329 |
+
"step": 99500
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 14.57,
|
| 1333 |
+
"learning_rate": 2.7134946079860102e-05,
|
| 1334 |
+
"loss": 0.0448,
|
| 1335 |
+
"step": 100000
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 14.65,
|
| 1339 |
+
"learning_rate": 2.67706208102594e-05,
|
| 1340 |
+
"loss": 0.0499,
|
| 1341 |
+
"step": 100500
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 14.72,
|
| 1345 |
+
"learning_rate": 2.64062955406587e-05,
|
| 1346 |
+
"loss": 0.0496,
|
| 1347 |
+
"step": 101000
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 14.79,
|
| 1351 |
+
"learning_rate": 2.6041970271058004e-05,
|
| 1352 |
+
"loss": 0.0507,
|
| 1353 |
+
"step": 101500
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 14.86,
|
| 1357 |
+
"learning_rate": 2.5677645001457303e-05,
|
| 1358 |
+
"loss": 0.051,
|
| 1359 |
+
"step": 102000
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 14.94,
|
| 1363 |
+
"learning_rate": 2.5313319731856604e-05,
|
| 1364 |
+
"loss": 0.051,
|
| 1365 |
+
"step": 102500
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 15.0,
|
| 1369 |
+
"eval_accuracy": 0.931511970074171,
|
| 1370 |
+
"eval_loss": 0.3619065284729004,
|
| 1371 |
+
"eval_runtime": 141.9378,
|
| 1372 |
+
"eval_samples_per_second": 32.232,
|
| 1373 |
+
"eval_steps_per_second": 5.376,
|
| 1374 |
+
"step": 102930
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 15.01,
|
| 1378 |
+
"learning_rate": 2.4948994462255903e-05,
|
| 1379 |
+
"loss": 0.0506,
|
| 1380 |
+
"step": 103000
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 15.08,
|
| 1384 |
+
"learning_rate": 2.4584669192655205e-05,
|
| 1385 |
+
"loss": 0.0419,
|
| 1386 |
+
"step": 103500
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 15.16,
|
| 1390 |
+
"learning_rate": 2.4220343923054503e-05,
|
| 1391 |
+
"loss": 0.0445,
|
| 1392 |
+
"step": 104000
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 15.23,
|
| 1396 |
+
"learning_rate": 2.3856018653453805e-05,
|
| 1397 |
+
"loss": 0.0449,
|
| 1398 |
+
"step": 104500
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 15.3,
|
| 1402 |
+
"learning_rate": 2.3491693383853107e-05,
|
| 1403 |
+
"loss": 0.0428,
|
| 1404 |
+
"step": 105000
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 15.37,
|
| 1408 |
+
"learning_rate": 2.3127368114252406e-05,
|
| 1409 |
+
"loss": 0.0474,
|
| 1410 |
+
"step": 105500
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 15.45,
|
| 1414 |
+
"learning_rate": 2.2763042844651704e-05,
|
| 1415 |
+
"loss": 0.0463,
|
| 1416 |
+
"step": 106000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 15.52,
|
| 1420 |
+
"learning_rate": 2.2398717575051006e-05,
|
| 1421 |
+
"loss": 0.0477,
|
| 1422 |
+
"step": 106500
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 15.59,
|
| 1426 |
+
"learning_rate": 2.2034392305450308e-05,
|
| 1427 |
+
"loss": 0.0459,
|
| 1428 |
+
"step": 107000
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 15.67,
|
| 1432 |
+
"learning_rate": 2.167006703584961e-05,
|
| 1433 |
+
"loss": 0.0505,
|
| 1434 |
+
"step": 107500
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 15.74,
|
| 1438 |
+
"learning_rate": 2.130574176624891e-05,
|
| 1439 |
+
"loss": 0.0445,
|
| 1440 |
+
"step": 108000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 15.81,
|
| 1444 |
+
"learning_rate": 2.0941416496648207e-05,
|
| 1445 |
+
"loss": 0.048,
|
| 1446 |
+
"step": 108500
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 15.88,
|
| 1450 |
+
"learning_rate": 2.057709122704751e-05,
|
| 1451 |
+
"loss": 0.046,
|
| 1452 |
+
"step": 109000
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 15.96,
|
| 1456 |
+
"learning_rate": 2.0212765957446807e-05,
|
| 1457 |
+
"loss": 0.0488,
|
| 1458 |
+
"step": 109500
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 16.0,
|
| 1462 |
+
"eval_accuracy": 0.932402742342834,
|
| 1463 |
+
"eval_loss": 0.3990395665168762,
|
| 1464 |
+
"eval_runtime": 141.9873,
|
| 1465 |
+
"eval_samples_per_second": 32.221,
|
| 1466 |
+
"eval_steps_per_second": 5.374,
|
| 1467 |
+
"step": 109792
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 16.03,
|
| 1471 |
+
"learning_rate": 1.984844068784611e-05,
|
| 1472 |
+
"loss": 0.0426,
|
| 1473 |
+
"step": 110000
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 16.1,
|
| 1477 |
+
"learning_rate": 1.948411541824541e-05,
|
| 1478 |
+
"loss": 0.0421,
|
| 1479 |
+
"step": 110500
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 16.18,
|
| 1483 |
+
"learning_rate": 1.9119790148644713e-05,
|
| 1484 |
+
"loss": 0.0431,
|
| 1485 |
+
"step": 111000
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 16.25,
|
| 1489 |
+
"learning_rate": 1.875546487904401e-05,
|
| 1490 |
+
"loss": 0.0417,
|
| 1491 |
+
"step": 111500
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 16.32,
|
| 1495 |
+
"learning_rate": 1.839113960944331e-05,
|
| 1496 |
+
"loss": 0.0393,
|
| 1497 |
+
"step": 112000
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 16.39,
|
| 1501 |
+
"learning_rate": 1.8026814339842612e-05,
|
| 1502 |
+
"loss": 0.0442,
|
| 1503 |
+
"step": 112500
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 16.47,
|
| 1507 |
+
"learning_rate": 1.7662489070241914e-05,
|
| 1508 |
+
"loss": 0.0435,
|
| 1509 |
+
"step": 113000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 16.54,
|
| 1513 |
+
"learning_rate": 1.7298163800641212e-05,
|
| 1514 |
+
"loss": 0.0462,
|
| 1515 |
+
"step": 113500
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 16.61,
|
| 1519 |
+
"learning_rate": 1.6933838531040514e-05,
|
| 1520 |
+
"loss": 0.0467,
|
| 1521 |
+
"step": 114000
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 16.69,
|
| 1525 |
+
"learning_rate": 1.6569513261439816e-05,
|
| 1526 |
+
"loss": 0.0437,
|
| 1527 |
+
"step": 114500
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 16.76,
|
| 1531 |
+
"learning_rate": 1.6205187991839114e-05,
|
| 1532 |
+
"loss": 0.046,
|
| 1533 |
+
"step": 115000
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 16.83,
|
| 1537 |
+
"learning_rate": 1.5840862722238413e-05,
|
| 1538 |
+
"loss": 0.0442,
|
| 1539 |
+
"step": 115500
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 16.9,
|
| 1543 |
+
"learning_rate": 1.5476537452637715e-05,
|
| 1544 |
+
"loss": 0.0435,
|
| 1545 |
+
"step": 116000
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 16.98,
|
| 1549 |
+
"learning_rate": 1.5112212183037017e-05,
|
| 1550 |
+
"loss": 0.0475,
|
| 1551 |
+
"step": 116500
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 17.0,
|
| 1555 |
+
"eval_accuracy": 0.9315557363417064,
|
| 1556 |
+
"eval_loss": 0.41956067085266113,
|
| 1557 |
+
"eval_runtime": 142.0163,
|
| 1558 |
+
"eval_samples_per_second": 32.215,
|
| 1559 |
+
"eval_steps_per_second": 5.373,
|
| 1560 |
+
"step": 116654
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 17.05,
|
| 1564 |
+
"learning_rate": 1.4747886913436319e-05,
|
| 1565 |
+
"loss": 0.0404,
|
| 1566 |
+
"step": 117000
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 17.12,
|
| 1570 |
+
"learning_rate": 1.4383561643835617e-05,
|
| 1571 |
+
"loss": 0.0419,
|
| 1572 |
+
"step": 117500
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 17.2,
|
| 1576 |
+
"learning_rate": 1.4019236374234917e-05,
|
| 1577 |
+
"loss": 0.0415,
|
| 1578 |
+
"step": 118000
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 17.27,
|
| 1582 |
+
"learning_rate": 1.365491110463422e-05,
|
| 1583 |
+
"loss": 0.0397,
|
| 1584 |
+
"step": 118500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 17.34,
|
| 1588 |
+
"learning_rate": 1.3290585835033518e-05,
|
| 1589 |
+
"loss": 0.0399,
|
| 1590 |
+
"step": 119000
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 17.41,
|
| 1594 |
+
"learning_rate": 1.2926260565432818e-05,
|
| 1595 |
+
"loss": 0.0438,
|
| 1596 |
+
"step": 119500
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 17.49,
|
| 1600 |
+
"learning_rate": 1.256193529583212e-05,
|
| 1601 |
+
"loss": 0.0453,
|
| 1602 |
+
"step": 120000
|
| 1603 |
+
},
|
| 1604 |
+
{
|
| 1605 |
+
"epoch": 17.56,
|
| 1606 |
+
"learning_rate": 1.219761002623142e-05,
|
| 1607 |
+
"loss": 0.0454,
|
| 1608 |
+
"step": 120500
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 17.63,
|
| 1612 |
+
"learning_rate": 1.183328475663072e-05,
|
| 1613 |
+
"loss": 0.0435,
|
| 1614 |
+
"step": 121000
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 17.71,
|
| 1618 |
+
"learning_rate": 1.146895948703002e-05,
|
| 1619 |
+
"loss": 0.0407,
|
| 1620 |
+
"step": 121500
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 17.78,
|
| 1624 |
+
"learning_rate": 1.1104634217429322e-05,
|
| 1625 |
+
"loss": 0.041,
|
| 1626 |
+
"step": 122000
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 17.85,
|
| 1630 |
+
"learning_rate": 1.074030894782862e-05,
|
| 1631 |
+
"loss": 0.0438,
|
| 1632 |
+
"step": 122500
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 17.92,
|
| 1636 |
+
"learning_rate": 1.0375983678227923e-05,
|
| 1637 |
+
"loss": 0.0449,
|
| 1638 |
+
"step": 123000
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 18.0,
|
| 1642 |
+
"learning_rate": 1.0011658408627223e-05,
|
| 1643 |
+
"loss": 0.0421,
|
| 1644 |
+
"step": 123500
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 18.0,
|
| 1648 |
+
"eval_accuracy": 0.9312395894209208,
|
| 1649 |
+
"eval_loss": 0.4355740249156952,
|
| 1650 |
+
"eval_runtime": 142.0074,
|
| 1651 |
+
"eval_samples_per_second": 32.217,
|
| 1652 |
+
"eval_steps_per_second": 5.373,
|
| 1653 |
+
"step": 123516
|
| 1654 |
+
}
|
| 1655 |
+
],
|
| 1656 |
+
"logging_steps": 500,
|
| 1657 |
+
"max_steps": 137240,
|
| 1658 |
+
"num_input_tokens_seen": 0,
|
| 1659 |
+
"num_train_epochs": 20,
|
| 1660 |
+
"save_steps": 500,
|
| 1661 |
+
"total_flos": 4.9509613574265683e+17,
|
| 1662 |
+
"train_batch_size": 6,
|
| 1663 |
+
"trial_name": null,
|
| 1664 |
+
"trial_params": null
|
| 1665 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:423bda2cced537c0b56430b36aae1d24bdcbfcf53ca95754403dfe30673cc7c9
|
| 3 |
+
size 4728
|
vocab.txt
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
<cls>
|
| 2 |
+
<pad>
|
| 3 |
+
<eos>
|
| 4 |
+
<unk>
|
| 5 |
+
L
|
| 6 |
+
A
|
| 7 |
+
G
|
| 8 |
+
V
|
| 9 |
+
S
|
| 10 |
+
E
|
| 11 |
+
R
|
| 12 |
+
T
|
| 13 |
+
I
|
| 14 |
+
D
|
| 15 |
+
P
|
| 16 |
+
K
|
| 17 |
+
Q
|
| 18 |
+
N
|
| 19 |
+
F
|
| 20 |
+
Y
|
| 21 |
+
M
|
| 22 |
+
H
|
| 23 |
+
W
|
| 24 |
+
C
|
| 25 |
+
X
|
| 26 |
+
B
|
| 27 |
+
U
|
| 28 |
+
Z
|
| 29 |
+
O
|
| 30 |
+
.
|
| 31 |
+
-
|
| 32 |
+
<null_1>
|
| 33 |
+
<mask>
|