Minbyul commited on
Commit
6bdc985
1 Parent(s): ad73942

End of training

Browse files
Files changed (4) hide show
  1. README.md +17 -1
  2. all_results.json +11 -11
  3. config.json +1 -1
  4. eval_results.json +11 -11
README.md CHANGED
@@ -2,10 +2,16 @@
2
  license: llama2
3
  base_model: Minbyul/meditron-7b-wo-medication_qa-sft
4
  tags:
 
 
 
 
5
  - trl
6
  - dpo
7
  - alignment-handbook
8
  - generated_from_trainer
 
 
9
  model-index:
10
  - name: meditron-7b-dpo-full-sft-wo-medication_qa
11
  results: []
@@ -16,7 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  # meditron-7b-dpo-full-sft-wo-medication_qa
18
 
19
- This model is a fine-tuned version of [Minbyul/meditron-7b-wo-medication_qa-sft](https://huggingface.co/Minbyul/meditron-7b-wo-medication_qa-sft) on an unknown dataset.
 
 
 
 
 
 
 
 
 
 
20
 
21
  ## Model description
22
 
 
2
  license: llama2
3
  base_model: Minbyul/meditron-7b-wo-medication_qa-sft
4
  tags:
5
+ - alignment-handbook
6
+ - trl
7
+ - dpo
8
+ - generated_from_trainer
9
  - trl
10
  - dpo
11
  - alignment-handbook
12
  - generated_from_trainer
13
+ datasets:
14
+ - HuggingFaceH4/ultrafeedback_binarized
15
  model-index:
16
  - name: meditron-7b-dpo-full-sft-wo-medication_qa
17
  results: []
 
22
 
23
  # meditron-7b-dpo-full-sft-wo-medication_qa
24
 
25
+ This model is a fine-tuned version of [Minbyul/meditron-7b-wo-medication_qa-sft](https://huggingface.co/Minbyul/meditron-7b-wo-medication_qa-sft) on the HuggingFaceH4/ultrafeedback_binarized dataset.
26
+ It achieves the following results on the evaluation set:
27
+ - Loss: 0.6347
28
+ - Rewards/chosen: -0.0755
29
+ - Rewards/rejected: -0.3042
30
+ - Rewards/accuracies: 0.7812
31
+ - Rewards/margins: 0.2287
32
+ - Logps/rejected: -613.1549
33
+ - Logps/chosen: -395.2336
34
+ - Logits/rejected: -1.0640
35
+ - Logits/chosen: -1.1745
36
 
37
  ## Model description
38
 
all_results.json CHANGED
@@ -1,18 +1,18 @@
1
  {
2
  "epoch": 0.99,
3
- "eval_logits/chosen": -1.1746665239334106,
4
- "eval_logits/rejected": -1.0633888244628906,
5
- "eval_logps/chosen": -394.94110107421875,
6
- "eval_logps/rejected": -613.10498046875,
7
- "eval_loss": 0.6342011094093323,
8
  "eval_rewards/accuracies": 0.78125,
9
- "eval_rewards/chosen": -0.07253288477659225,
10
- "eval_rewards/margins": 0.23117530345916748,
11
- "eval_rewards/rejected": -0.30370816588401794,
12
- "eval_runtime": 30.5473,
13
  "eval_samples": 240,
14
- "eval_samples_per_second": 7.857,
15
- "eval_steps_per_second": 0.262,
16
  "train_loss": 0.5081947011462713,
17
  "train_runtime": 896.7003,
18
  "train_samples": 3790,
 
1
  {
2
  "epoch": 0.99,
3
+ "eval_logits/chosen": -1.1745316982269287,
4
+ "eval_logits/rejected": -1.0640313625335693,
5
+ "eval_logps/chosen": -395.233642578125,
6
+ "eval_logps/rejected": -613.1549072265625,
7
+ "eval_loss": 0.6347063779830933,
8
  "eval_rewards/accuracies": 0.78125,
9
+ "eval_rewards/chosen": -0.0754583477973938,
10
+ "eval_rewards/margins": 0.22874917089939117,
11
+ "eval_rewards/rejected": -0.3042075037956238,
12
+ "eval_runtime": 30.2202,
13
  "eval_samples": 240,
14
+ "eval_samples_per_second": 7.942,
15
+ "eval_steps_per_second": 0.265,
16
  "train_loss": 0.5081947011462713,
17
  "train_runtime": 896.7003,
18
  "train_samples": 3790,
config.json CHANGED
@@ -23,6 +23,6 @@
23
  "tie_word_embeddings": false,
24
  "torch_dtype": "bfloat16",
25
  "transformers_version": "4.39.0.dev0",
26
- "use_cache": false,
27
  "vocab_size": 32017
28
  }
 
23
  "tie_word_embeddings": false,
24
  "torch_dtype": "bfloat16",
25
  "transformers_version": "4.39.0.dev0",
26
+ "use_cache": true,
27
  "vocab_size": 32017
28
  }
eval_results.json CHANGED
@@ -1,16 +1,16 @@
1
  {
2
  "epoch": 0.99,
3
- "eval_logits/chosen": -1.1746665239334106,
4
- "eval_logits/rejected": -1.0633888244628906,
5
- "eval_logps/chosen": -394.94110107421875,
6
- "eval_logps/rejected": -613.10498046875,
7
- "eval_loss": 0.6342011094093323,
8
  "eval_rewards/accuracies": 0.78125,
9
- "eval_rewards/chosen": -0.07253288477659225,
10
- "eval_rewards/margins": 0.23117530345916748,
11
- "eval_rewards/rejected": -0.30370816588401794,
12
- "eval_runtime": 30.5473,
13
  "eval_samples": 240,
14
- "eval_samples_per_second": 7.857,
15
- "eval_steps_per_second": 0.262
16
  }
 
1
  {
2
  "epoch": 0.99,
3
+ "eval_logits/chosen": -1.1745316982269287,
4
+ "eval_logits/rejected": -1.0640313625335693,
5
+ "eval_logps/chosen": -395.233642578125,
6
+ "eval_logps/rejected": -613.1549072265625,
7
+ "eval_loss": 0.6347063779830933,
8
  "eval_rewards/accuracies": 0.78125,
9
+ "eval_rewards/chosen": -0.0754583477973938,
10
+ "eval_rewards/margins": 0.22874917089939117,
11
+ "eval_rewards/rejected": -0.3042075037956238,
12
+ "eval_runtime": 30.2202,
13
  "eval_samples": 240,
14
+ "eval_samples_per_second": 7.942,
15
+ "eval_steps_per_second": 0.265
16
  }