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
b7fbd0b
•
1 Parent(s): b209552

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_2.0.json +1 -0
  6. metrics_epoch_2.0_fold_0_lr_1e-05_seed_5678_weight_2.0.json +1 -0
  7. results_epoch_0.96_fold_0_lr_1e-05_seed_5678_weight_2.0.json +0 -0
  8. results_epoch_2.0_fold_0_lr_1e-05_seed_5678_weight_2.0.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: 0.6719
22
- - Eval/rewards/chosen: 3.3033
23
- - Eval/logps/chosen: -167.4968
24
- - Eval/rewards/rejected: 2.5694
25
- - Eval/logps/rejected: -214.6203
26
- - Eval/rewards/margins: 0.7339
27
- - Eval/kl: 27.9394
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
- | 0.5643 | 0.96 | 12 | 0.6616 | 20.6058 |
62
- | 0.5188 | 2.0 | 25 | 0.6719 | 27.9394 |
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: 0.6739
22
+ - Eval/rewards/chosen: 0.1558
23
+ - Eval/logps/chosen: -198.9717
24
+ - Eval/rewards/rejected: 0.1376
25
+ - Eval/logps/rejected: -238.9384
26
+ - Eval/rewards/margins: 0.0182
27
+ - Eval/kl: 1.4137
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
+ | 0.5651 | 0.96 | 12 | 0.6738 | 0.3319 |
62
+ | 0.6129 | 2.0 | 25 | 0.6739 | 1.4137 |
63
 
64
 
65
  ### Framework versions
adapter_config.json CHANGED
@@ -20,10 +20,10 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
- "v_proj",
24
  "q_proj",
25
- "k_proj",
26
- "o_proj"
 
27
  ],
28
  "task_type": "CAUSAL_LM",
29
  "use_dora": false,
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
23
  "q_proj",
24
+ "v_proj",
25
+ "o_proj",
26
+ "k_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:a4527ca7d2c9e75f611647772f1c417097f415a07ebe3f41a2599123982f932f
3
  size 27297544
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd5bacd8f7d04ec41f7a5bdd5d86ee9b42c4e8a9f59c3fb3da27bbd600e74607
3
  size 27297544
metrics.jsonl CHANGED
@@ -14,3 +14,5 @@
14
  {"epoch": 2.0, "precision": 0.624999998046875, "recall": 0.6666666644444444, "fold": 0}
15
  {"epoch": 0.96, "precision": 0.3749999994791667, "recall": 0.899999997, "fold": 0}
16
  {"epoch": 2.0, "precision": 0.9090909049586776, "recall": 0.6666666644444444, "fold": 0}
 
 
 
14
  {"epoch": 2.0, "precision": 0.624999998046875, "recall": 0.6666666644444444, "fold": 0}
15
  {"epoch": 0.96, "precision": 0.3749999994791667, "recall": 0.899999997, "fold": 0}
16
  {"epoch": 2.0, "precision": 0.9090909049586776, "recall": 0.6666666644444444, "fold": 0}
17
+ {"epoch": 0.96, "precision": 0.35897435851413545, "recall": 0.9333333302222222, "fold": 0}
18
+ {"epoch": 2.0, "precision": 0.3766233761342554, "recall": 0.9666666634444444, "fold": 0}
metrics_epoch_0.96_fold_0_lr_1e-05_seed_5678_weight_2.0.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"epoch": 0.96, "precision": 0.35897435851413545, "recall": 0.9333333302222222, "fold": 0}
metrics_epoch_2.0_fold_0_lr_1e-05_seed_5678_weight_2.0.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"epoch": 2.0, "precision": 0.3766233761342554, "recall": 0.9666666634444444, "fold": 0}
results_epoch_0.96_fold_0_lr_1e-05_seed_5678_weight_2.0.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_2.0.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 sha256:424b754ebd47e29ce17ae5e53d89c6ab439b604a77dbde60473874ff30be8777
3
  size 5688
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a03251dfea3eb69937766de860253fbf86f9aaaa4e35d3809f42e2dc7ca0496
3
  size 5688