End of training
Browse files- README.md +18 -1
- all_results.json +13 -0
- config.json +1 -1
- eval_results.json +16 -0
- runs/Oct05_11-54-20_68660d2c5c1b/events.out.tfevents.1728131559.68660d2c5c1b.2665.1 +3 -0
README.md
CHANGED
@@ -3,9 +3,16 @@ library_name: transformers
|
|
3 |
license: apache-2.0
|
4 |
base_model: pbevan11/Mistral-Nemo-Instruct-MCAI-SFT-revision-only
|
5 |
tags:
|
|
|
6 |
- trl
|
7 |
- dpo
|
8 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
model-index:
|
10 |
- name: Mistral-Nemo-Instruct-MCAI-SFT-DPO-revision-only
|
11 |
results: []
|
@@ -16,7 +23,17 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
|
17 |
# Mistral-Nemo-Instruct-MCAI-SFT-DPO-revision-only
|
18 |
|
19 |
-
This model is a fine-tuned version of [pbevan11/Mistral-Nemo-Instruct-MCAI-SFT-revision-only](https://huggingface.co/pbevan11/Mistral-Nemo-Instruct-MCAI-SFT-revision-only) on the
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
20 |
|
21 |
## Model description
|
22 |
|
|
|
3 |
license: apache-2.0
|
4 |
base_model: pbevan11/Mistral-Nemo-Instruct-MCAI-SFT-revision-only
|
5 |
tags:
|
6 |
+
- alignment-handbook
|
7 |
- trl
|
8 |
- dpo
|
9 |
- generated_from_trainer
|
10 |
+
- trl
|
11 |
+
- dpo
|
12 |
+
- generated_from_trainer
|
13 |
+
datasets:
|
14 |
+
- pbevan11/multilingual-constitutional-preference-pairs-revision-only
|
15 |
+
- pbevan11/ultrafeedback_binarized_multilingual
|
16 |
model-index:
|
17 |
- name: Mistral-Nemo-Instruct-MCAI-SFT-DPO-revision-only
|
18 |
results: []
|
|
|
23 |
|
24 |
# Mistral-Nemo-Instruct-MCAI-SFT-DPO-revision-only
|
25 |
|
26 |
+
This model is a fine-tuned version of [pbevan11/Mistral-Nemo-Instruct-MCAI-SFT-revision-only](https://huggingface.co/pbevan11/Mistral-Nemo-Instruct-MCAI-SFT-revision-only) on the pbevan11/multilingual-constitutional-preference-pairs-revision-only and the pbevan11/ultrafeedback_binarized_multilingual datasets.
|
27 |
+
It achieves the following results on the evaluation set:
|
28 |
+
- Loss: 0.5140
|
29 |
+
- Rewards/chosen: -0.6440
|
30 |
+
- Rewards/rejected: -1.3117
|
31 |
+
- Rewards/accuracies: 0.6824
|
32 |
+
- Rewards/margins: 0.6678
|
33 |
+
- Logps/rejected: -215.4680
|
34 |
+
- Logps/chosen: -196.1248
|
35 |
+
- Logits/rejected: -0.0511
|
36 |
+
- Logits/chosen: -0.0322
|
37 |
|
38 |
## Model description
|
39 |
|
all_results.json
CHANGED
@@ -1,5 +1,18 @@
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
"total_flos": 0.0,
|
4 |
"train_loss": 0.5670995542988975,
|
5 |
"train_runtime": 1780.4141,
|
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
+
"eval_logits/chosen": -0.03224845603108406,
|
4 |
+
"eval_logits/rejected": -0.051055099815130234,
|
5 |
+
"eval_logps/chosen": -196.12477111816406,
|
6 |
+
"eval_logps/rejected": -215.46804809570312,
|
7 |
+
"eval_loss": 0.5139751434326172,
|
8 |
+
"eval_rewards/accuracies": 0.6824324131011963,
|
9 |
+
"eval_rewards/chosen": -0.6439886093139648,
|
10 |
+
"eval_rewards/margins": 0.6677516102790833,
|
11 |
+
"eval_rewards/rejected": -1.3117402791976929,
|
12 |
+
"eval_runtime": 216.2859,
|
13 |
+
"eval_samples": 2341,
|
14 |
+
"eval_samples_per_second": 10.824,
|
15 |
+
"eval_steps_per_second": 0.171,
|
16 |
"total_flos": 0.0,
|
17 |
"train_loss": 0.5670995542988975,
|
18 |
"train_runtime": 1780.4141,
|
config.json
CHANGED
@@ -22,6 +22,6 @@
|
|
22 |
"tie_word_embeddings": false,
|
23 |
"torch_dtype": "bfloat16",
|
24 |
"transformers_version": "4.45.1",
|
25 |
-
"use_cache":
|
26 |
"vocab_size": 131072
|
27 |
}
|
|
|
22 |
"tie_word_embeddings": false,
|
23 |
"torch_dtype": "bfloat16",
|
24 |
"transformers_version": "4.45.1",
|
25 |
+
"use_cache": true,
|
26 |
"vocab_size": 131072
|
27 |
}
|
eval_results.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_logits/chosen": -0.03224845603108406,
|
4 |
+
"eval_logits/rejected": -0.051055099815130234,
|
5 |
+
"eval_logps/chosen": -196.12477111816406,
|
6 |
+
"eval_logps/rejected": -215.46804809570312,
|
7 |
+
"eval_loss": 0.5139751434326172,
|
8 |
+
"eval_rewards/accuracies": 0.6824324131011963,
|
9 |
+
"eval_rewards/chosen": -0.6439886093139648,
|
10 |
+
"eval_rewards/margins": 0.6677516102790833,
|
11 |
+
"eval_rewards/rejected": -1.3117402791976929,
|
12 |
+
"eval_runtime": 216.2859,
|
13 |
+
"eval_samples": 2341,
|
14 |
+
"eval_samples_per_second": 10.824,
|
15 |
+
"eval_steps_per_second": 0.171
|
16 |
+
}
|
runs/Oct05_11-54-20_68660d2c5c1b/events.out.tfevents.1728131559.68660d2c5c1b.2665.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b93ed606d70a40984dcd9bf4cab55b5b65610e9e21aceabb10e50d2e468389ca
|
3 |
+
size 828
|