BraylonDash commited on
Commit
4cf58dc
1 Parent(s): 090cb4a

Model save

Browse files
README.md CHANGED
@@ -2,14 +2,10 @@
2
  license: mit
3
  library_name: peft
4
  tags:
5
- - alignment-handbook
6
- - generated_from_trainer
7
  - trl
8
  - dpo
9
  - generated_from_trainer
10
  base_model: microsoft/phi-2
11
- datasets:
12
- - HuggingFaceH4/ultrafeedback_binarized
13
  model-index:
14
  - name: phi-2-gpo-test-iter-1
15
  results: []
@@ -20,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  # phi-2-gpo-test-iter-1
22
 
23
- This model is a fine-tuned version of [BraylonDash/phi-2-gpo-test-iter-1](https://huggingface.co/BraylonDash/phi-2-gpo-test-iter-1) on the HuggingFaceH4/ultrafeedback_binarized dataset.
24
 
25
  ## Model description
26
 
 
2
  license: mit
3
  library_name: peft
4
  tags:
 
 
5
  - trl
6
  - dpo
7
  - generated_from_trainer
8
  base_model: microsoft/phi-2
 
 
9
  model-index:
10
  - name: phi-2-gpo-test-iter-1
11
  results: []
 
16
 
17
  # phi-2-gpo-test-iter-1
18
 
19
+ This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on the None dataset.
20
 
21
  ## Model description
22
 
all_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "epoch": 2.0,
3
- "train_loss": 0.0009918120922520757,
4
- "train_runtime": 3.8679,
5
  "train_samples": 30567,
6
- "train_samples_per_second": 1.034,
7
- "train_steps_per_second": 0.517
8
  }
 
1
  {
2
  "epoch": 2.0,
3
+ "train_loss": 0.0008448870503343642,
4
+ "train_runtime": 3.5318,
5
  "train_samples": 30567,
6
+ "train_samples_per_second": 2.265,
7
+ "train_steps_per_second": 0.566
8
  }
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "epoch": 2.0,
3
- "train_loss": 0.0009918120922520757,
4
- "train_runtime": 3.8679,
5
  "train_samples": 30567,
6
- "train_samples_per_second": 1.034,
7
- "train_steps_per_second": 0.517
8
  }
 
1
  {
2
  "epoch": 2.0,
3
+ "train_loss": 0.0008448870503343642,
4
+ "train_runtime": 3.5318,
5
  "train_samples": 30567,
6
+ "train_samples_per_second": 2.265,
7
+ "train_steps_per_second": 0.566
8
  }
trainer_state.json CHANGED
@@ -11,11 +11,11 @@
11
  {
12
  "epoch": 1.0,
13
  "learning_rate": 5e-06,
14
- "logits/chosen": -0.04497978091239929,
15
- "logits/rejected": 0.10493635386228561,
16
- "logps/chosen": -188.7808074951172,
17
- "logps/rejected": -71.98552703857422,
18
- "loss": 0.001,
19
  "rewards/accuracies": 0.0,
20
  "rewards/chosen": 0.0,
21
  "rewards/margins": 0.0,
@@ -26,10 +26,10 @@
26
  "epoch": 2.0,
27
  "step": 2,
28
  "total_flos": 0.0,
29
- "train_loss": 0.0009918120922520757,
30
- "train_runtime": 3.8679,
31
- "train_samples_per_second": 1.034,
32
- "train_steps_per_second": 0.517
33
  }
34
  ],
35
  "logging_steps": 10,
 
11
  {
12
  "epoch": 1.0,
13
  "learning_rate": 5e-06,
14
+ "logits/chosen": 0.09949145466089249,
15
+ "logits/rejected": -0.044053155928850174,
16
+ "logps/chosen": -259.9786682128906,
17
+ "logps/rejected": -270.3120422363281,
18
+ "loss": 0.0008,
19
  "rewards/accuracies": 0.0,
20
  "rewards/chosen": 0.0,
21
  "rewards/margins": 0.0,
 
26
  "epoch": 2.0,
27
  "step": 2,
28
  "total_flos": 0.0,
29
+ "train_loss": 0.0008448870503343642,
30
+ "train_runtime": 3.5318,
31
+ "train_samples_per_second": 2.265,
32
+ "train_steps_per_second": 0.566
33
  }
34
  ],
35
  "logging_steps": 10,
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8920f751d3a9bbe064c39e1cee59d95a52dfb1d70c1cd72878db0ab77d2ed09d
3
- size 5880
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9ece1ca5802cca4653ce999493d4cb39e82f243e1ce3e83511f56e65dabaa96
3
+ size 5816