Hugging Face's logo

PaulD
/
llama3_false_positives_1010_KTO_hp_screening_seeds

PEFT
Safetensors
trl
kto
Generated from Trainer
Model card Files Community
PaulD commited on
Commit
e4829c4
•
1 Parent(s): a59dc80

End of training

Browse files
Files changed (9) hide show
  1. README.md +12 -12
  2. adapter_config.json +3 -3
  3. adapter_model.safetensors +1 -1
  4. metrics.jsonl +2 -0
  5. metrics_epoch_0.96_fold_0_lr_1e-05_seed_5678_weight_5.0.json +1 -0
  6. metrics_epoch_2.0_fold_0_lr_1e-05_seed_5678_weight_5.0.json +1 -0
  7. results_epoch_0.96_fold_0_lr_1e-05_seed_5678_weight_5.0.json +0 -0
  8. results_epoch_2.0_fold_0_lr_1e-05_seed_5678_weight_5.0.json +0 -0
  9. training_args.bin +1 -1
README.md CHANGED
@@ -18,13 +18,13 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 1.1232
22
- - Eval/rewards/chosen: 1.9522
23
- - Eval/logps/chosen: -181.0075
24
- - Eval/rewards/rejected: 1.3331
25
- - Eval/logps/rejected: -226.9831
26
- - Eval/rewards/margins: 0.6191
27
- - Eval/kl: 18.6818
28
 
29
  ## Model description
30
 
@@ -43,7 +43,7 @@ More information needed
43
  ### Training hyperparameters
44
 
45
  The following hyperparameters were used during training:
46
- - learning_rate: 0.0001
47
  - train_batch_size: 1
48
  - eval_batch_size: 2
49
  - seed: 5678
@@ -56,10 +56,10 @@ The following hyperparameters were used during training:
56
 
57
  ### Training results
58
 
59
- | Training Loss | Epoch | Step | Validation Loss | |
60
- |:-------------:|:-----:|:----:|:---------------:|:-------:|
61
- | 0.6672 | 0.96 | 12 | 0.8960 | 0.0129 |
62
- | 0.7023 | 2.0 | 25 | 1.1232 | 18.6818 |
63
 
64
 
65
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 1.1781
22
+ - Eval/rewards/chosen: -0.0396
23
+ - Eval/logps/chosen: -200.9254
24
+ - Eval/rewards/rejected: -0.0766
25
+ - Eval/logps/rejected: -241.0806
26
+ - Eval/rewards/margins: 0.0370
27
+ - Eval/kl: 0.0308
28
 
29
  ## Model description
30
 
 
43
  ### Training hyperparameters
44
 
45
  The following hyperparameters were used during training:
46
+ - learning_rate: 1e-05
47
  - train_batch_size: 1
48
  - eval_batch_size: 2
49
  - seed: 5678
 
56
 
57
  ### Training results
58
 
59
+ | Training Loss | Epoch | Step | Validation Loss | |
60
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
61
+ | 0.7403 | 0.96 | 12 | 1.2009 | 0.1233 |
62
+ | 0.961 | 2.0 | 25 | 1.1781 | 0.0308 |
63
 
64
 
65
  ### Framework versions
adapter_config.json CHANGED
@@ -20,10 +20,10 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
- "o_proj",
24
- "v_proj",
25
  "q_proj",
26
- "k_proj"
 
27
  ],
28
  "task_type": "CAUSAL_LM",
29
  "use_dora": false,
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
+ "k_proj",
 
24
  "q_proj",
25
+ "v_proj",
26
+ "o_proj"
27
  ],
28
  "task_type": "CAUSAL_LM",
29
  "use_dora": false,
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5e906d45412e0e081df58bfb409ec0a438a1ff383a0a21e37a2e3a89259033ae
3
  size 27297544
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd6946db558e74bfb454bad307b0bdc33704e39f257b512a5d6be81e58304d46
3
  size 27297544
metrics.jsonl CHANGED
@@ -20,3 +20,5 @@
20
  {"epoch": 2.0, "precision": 0.2238805968478503, "recall": 0.9999999966666666, "fold": 0}
21
  {"epoch": 0.96, "precision": 0.49122806931363494, "recall": 0.9333333302222222, "fold": 0}
22
  {"epoch": 2.0, "precision": 0.6444444430123457, "recall": 0.9666666634444444, "fold": 0}
 
 
 
20
  {"epoch": 2.0, "precision": 0.2238805968478503, "recall": 0.9999999966666666, "fold": 0}
21
  {"epoch": 0.96, "precision": 0.49122806931363494, "recall": 0.9333333302222222, "fold": 0}
22
  {"epoch": 2.0, "precision": 0.6444444430123457, "recall": 0.9666666634444444, "fold": 0}
23
+ {"epoch": 0.96, "precision": 0.34939758994048487, "recall": 0.9666666634444444, "fold": 0}
24
+ {"epoch": 2.0, "precision": 0.3717948713182117, "recall": 0.9666666634444444, "fold": 0}
metrics_epoch_0.96_fold_0_lr_1e-05_seed_5678_weight_5.0.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"epoch": 0.96, "precision": 0.34939758994048487, "recall": 0.9666666634444444, "fold": 0}
metrics_epoch_2.0_fold_0_lr_1e-05_seed_5678_weight_5.0.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"epoch": 2.0, "precision": 0.3717948713182117, "recall": 0.9666666634444444, "fold": 0}
results_epoch_0.96_fold_0_lr_1e-05_seed_5678_weight_5.0.json ADDED
The diff for this file is too large to render. See raw diff
 
results_epoch_2.0_fold_0_lr_1e-05_seed_5678_weight_5.0.json ADDED
The diff for this file is too large to render. See raw diff
 
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e21138be0fdf7f83c5156b0c3de984b0c18bb1857dfa4a84fed5074bbaf1482e
3
  size 5688
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5783f3b3aa05b56860198428463dd5f7822af851e3ab42375ceb413fc156243e
3
  size 5688