Hugging Face's logo

aritrasen
/
mistral-dpo

PEFT
TensorBoard
Safetensors
Generated from Trainer
Model card Files Metrics Community
aritrasen commited on
Commit
0782eaf
•
1 Parent(s): b320a23

aritrasen/mistral-dpo

Browse files
Files changed (4) hide show
  1. README.md +34 -12
  2. adapter_model.safetensors +1 -1
  3. runs/Dec21_18-30-01_95c7d1f2a3f6/events.out.tfevents.1703183408.95c7d1f2a3f6.26.0 +3 -0
  4. training_args.bin +1 -1
README.md CHANGED
@@ -7,8 +7,6 @@ base_model: TheBloke/OpenHermes-2-Mistral-7B-GPTQ
7
  model-index:
8
  - name: mistral-dpo
9
  results: []
10
- datasets:
11
- - Dahoas/full-hh-rlhf
12
  ---
13
 
14
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -18,15 +16,15 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [TheBloke/OpenHermes-2-Mistral-7B-GPTQ](https://huggingface.co/TheBloke/OpenHermes-2-Mistral-7B-GPTQ) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.6930
22
- - Rewards/chosen: -0.0558
23
- - Rewards/rejected: -0.0669
24
  - Rewards/accuracies: 0.5096
25
- - Rewards/margins: 0.0111
26
- - Logps/rejected: -200.9969
27
- - Logps/chosen: -193.8321
28
- - Logits/rejected: -2.2922
29
- - Logits/chosen: -2.3906
30
 
31
  ## Model description
32
 
@@ -52,14 +50,38 @@ The following hyperparameters were used during training:
52
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
53
  - lr_scheduler_type: linear
54
  - lr_scheduler_warmup_steps: 2
55
- - training_steps: 10
56
  - mixed_precision_training: Native AMP
57
 
58
  ### Training results
59
 
60
  | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
61
  |:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
62
- | 0.6825 | 0.0 | 10 | 0.6930 | -0.0558 | -0.0669 | 0.5096 | 0.0111 | -200.9969 | -193.8321 | -2.2922 | -2.3906 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
63
 
64
 
65
  ### Framework versions
 
7
  model-index:
8
  - name: mistral-dpo
9
  results: []
 
 
10
  ---
11
 
12
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
16
 
17
  This model is a fine-tuned version of [TheBloke/OpenHermes-2-Mistral-7B-GPTQ](https://huggingface.co/TheBloke/OpenHermes-2-Mistral-7B-GPTQ) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.8911
20
+ - Rewards/chosen: 0.5387
21
+ - Rewards/rejected: 0.4878
22
  - Rewards/accuracies: 0.5096
23
+ - Rewards/margins: 0.0509
24
+ - Logps/rejected: -174.3804
25
+ - Logps/chosen: -178.5185
26
+ - Logits/rejected: -2.5028
27
+ - Logits/chosen: -2.5350
28
 
29
  ## Model description
30
 
 
50
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
  - lr_scheduler_type: linear
52
  - lr_scheduler_warmup_steps: 2
53
+ - training_steps: 250
54
  - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
59
  |:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
60
+ | 0.6703 | 0.0 | 10 | 0.6842 | -0.0001 | -0.0268 | 0.5865 | 0.0267 | -179.5257 | -183.9063 | -2.4290 | -2.4720 |
61
+ | 0.7119 | 0.0 | 20 | 0.6751 | 0.1584 | 0.0990 | 0.5769 | 0.0594 | -178.2678 | -182.3211 | -2.4542 | -2.4988 |
62
+ | 0.647 | 0.0 | 30 | 0.6702 | 0.3569 | 0.2540 | 0.5769 | 0.1029 | -176.7180 | -180.3367 | -2.4886 | -2.5306 |
63
+ | 0.6748 | 0.0 | 40 | 0.6712 | 0.3439 | 0.2229 | 0.5288 | 0.1210 | -177.0292 | -180.4664 | -2.5206 | -2.5581 |
64
+ | 0.6513 | 0.0 | 50 | 0.6707 | 0.4403 | 0.2838 | 0.5577 | 0.1565 | -176.4200 | -179.5021 | -2.5608 | -2.5853 |
65
+ | 0.6103 | 0.0 | 60 | 0.6695 | 0.6831 | 0.4769 | 0.5577 | 0.2063 | -174.4892 | -177.0740 | -2.5719 | -2.5933 |
66
+ | 1.0313 | 0.01 | 70 | 0.6724 | 0.7062 | 0.5084 | 0.5577 | 0.1978 | -174.1739 | -176.8436 | -2.5543 | -2.5843 |
67
+ | 0.6876 | 0.01 | 80 | 0.6804 | 0.6995 | 0.5144 | 0.5385 | 0.1850 | -174.1135 | -176.9104 | -2.5443 | -2.5829 |
68
+ | 0.9661 | 0.01 | 90 | 0.6828 | 0.7118 | 0.5376 | 0.5385 | 0.1742 | -173.8821 | -176.7873 | -2.5479 | -2.5846 |
69
+ | 0.7354 | 0.01 | 100 | 0.6757 | 0.6765 | 0.5039 | 0.5577 | 0.1726 | -174.2186 | -177.1401 | -2.5399 | -2.5758 |
70
+ | 1.0127 | 0.01 | 110 | 0.7129 | 0.6089 | 0.4855 | 0.5288 | 0.1234 | -174.4033 | -177.8165 | -2.5464 | -2.5760 |
71
+ | 1.0366 | 0.01 | 120 | 0.7440 | 0.6068 | 0.4946 | 0.5481 | 0.1122 | -174.3115 | -177.8369 | -2.5516 | -2.5804 |
72
+ | 1.2145 | 0.01 | 130 | 0.7564 | 0.6521 | 0.5396 | 0.5673 | 0.1125 | -173.8620 | -177.3846 | -2.5608 | -2.5878 |
73
+ | 0.8342 | 0.01 | 140 | 0.7649 | 0.6639 | 0.5519 | 0.5385 | 0.1119 | -173.7388 | -177.2668 | -2.5547 | -2.5828 |
74
+ | 0.7402 | 0.01 | 150 | 0.7991 | 0.5831 | 0.4883 | 0.5 | 0.0948 | -174.3747 | -178.0745 | -2.5498 | -2.5775 |
75
+ | 0.7162 | 0.01 | 160 | 0.8396 | 0.6134 | 0.5474 | 0.5096 | 0.0659 | -173.7835 | -177.7718 | -2.5445 | -2.5713 |
76
+ | 0.9396 | 0.01 | 170 | 0.8573 | 0.5700 | 0.5144 | 0.5288 | 0.0556 | -174.1144 | -178.2057 | -2.5326 | -2.5629 |
77
+ | 0.5958 | 0.01 | 180 | 0.8708 | 0.5526 | 0.5017 | 0.5288 | 0.0509 | -174.2406 | -178.3789 | -2.5227 | -2.5540 |
78
+ | 0.7588 | 0.02 | 190 | 0.8865 | 0.5428 | 0.4977 | 0.5288 | 0.0450 | -174.2806 | -178.4775 | -2.5207 | -2.5493 |
79
+ | 0.7811 | 0.02 | 200 | 0.8933 | 0.5797 | 0.5429 | 0.5192 | 0.0368 | -173.8286 | -178.1080 | -2.5171 | -2.5434 |
80
+ | 0.5735 | 0.02 | 210 | 0.8907 | 0.5577 | 0.5174 | 0.5288 | 0.0403 | -174.0838 | -178.3279 | -2.5069 | -2.5366 |
81
+ | 0.7709 | 0.02 | 220 | 0.8886 | 0.5602 | 0.5167 | 0.5192 | 0.0435 | -174.0907 | -178.3035 | -2.5041 | -2.5361 |
82
+ | 0.4914 | 0.02 | 230 | 0.8884 | 0.5237 | 0.4766 | 0.5192 | 0.0471 | -174.4924 | -178.6684 | -2.5050 | -2.5375 |
83
+ | 0.739 | 0.02 | 240 | 0.8910 | 0.5281 | 0.4796 | 0.5192 | 0.0485 | -174.4621 | -178.6240 | -2.5027 | -2.5351 |
84
+ | 0.5743 | 0.02 | 250 | 0.8911 | 0.5387 | 0.4878 | 0.5096 | 0.0509 | -174.3804 | -178.5185 | -2.5028 | -2.5350 |
85
 
86
 
87
  ### Framework versions
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3a7670b20147de176a9b407e70da7aa40acf3479f3f93ba737129a62c21535cf
3
  size 6832600
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ee1512f5d20c94e2d685820ce34546d555489cb396373434b6e54d8ba4df875
3
  size 6832600
runs/Dec21_18-30-01_95c7d1f2a3f6/events.out.tfevents.1703183408.95c7d1f2a3f6.26.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f95051c65615cf007739a26f224d4cae371399059016e716fde064d51b009d1f
3
+ size 40047
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d1a5c488c6ab1fd32c0bcdf7d0081ef6565e6436f00739c2c4d5781c702e6ec7
3
  size 4219
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a744ecc30871a71a7fc5ed07942856a6e759e716fa13531c07b20ed7930a0f2e
3
  size 4219