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
c3bf95c
·
verified ·
1 Parent(s): 401243b

End of training

Browse files
Files changed (11) hide show
  1. README.md +13 -13
  2. adapter_config.json +1 -1
  3. adapter_model.safetensors +1 -1
  4. metrics.jsonl +3 -0
  5. metrics_epoch_0.9411764705882353_fold_0_lr_0.0001_seed_9012_weight_5.0.json +1 -0
  6. metrics_epoch_1.9607843137254903_fold_0_lr_0.0001_seed_9012_weight_5.0.json +1 -0
  7. metrics_epoch_2.980392156862745_fold_0_lr_0.0001_seed_9012_weight_5.0.json +1 -0
  8. results_epoch_0.9411764705882353_fold_0_lr_0.0001_seed_9012_weight_5.0.json +0 -0
  9. results_epoch_1.9607843137254903_fold_0_lr_0.0001_seed_9012_weight_5.0.json +0 -0
  10. results_epoch_2.980392156862745_fold_0_lr_0.0001_seed_9012_weight_5.0.json +0 -0
  11. 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: 0.5882
22
- - Eval/rewards/chosen: -241.2744
23
- - Eval/logps/chosen: -2623.3453
24
- - Eval/rewards/rejected: -268.1879
25
- - Eval/logps/rejected: -2915.7442
26
- - Eval/rewards/margins: 26.9135
27
- - Eval/kl: 0.0
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: 0.001
47
  - train_batch_size: 1
48
  - eval_batch_size: 2
49
  - seed: 9012
@@ -56,11 +56,11 @@ The following hyperparameters were used during training:
56
 
57
  ### Training results
58
 
59
- | Training Loss | Epoch | Step | Validation Loss | |
60
- |:-------------:|:------:|:----:|:---------------:|:---:|
61
- | 0.875 | 0.9412 | 12 | 0.5882 | 0.0 |
62
- | 0.875 | 1.9608 | 25 | 0.5882 | 0.0 |
63
- | 0.875 | 2.9804 | 38 | 0.5882 | 0.0 |
64
 
65
 
66
  ### 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: 1.0826
22
+ - Eval/rewards/chosen: 2.6488
23
+ - Eval/logps/chosen: -184.1138
24
+ - Eval/rewards/rejected: 1.0242
25
+ - Eval/logps/rejected: -223.6231
26
+ - Eval/rewards/margins: 1.6246
27
+ - Eval/kl: 17.0687
28
 
29
  ## Model description
30
 
 
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: 9012
 
56
 
57
  ### Training results
58
 
59
+ | Training Loss | Epoch | Step | Validation Loss | |
60
+ |:-------------:|:------:|:----:|:---------------:|:-------:|
61
+ | 0.7159 | 0.9412 | 12 | 0.9105 | 0.0 |
62
+ | 0.4236 | 1.9608 | 25 | 1.0808 | 11.7436 |
63
+ | 0.2641 | 2.9804 | 38 | 1.0826 | 17.0687 |
64
 
65
 
66
  ### Framework versions
adapter_config.json CHANGED
@@ -20,8 +20,8 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
- "o_proj",
24
  "k_proj",
 
25
  "v_proj",
26
  "q_proj"
27
  ],
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
23
  "k_proj",
24
+ "o_proj",
25
  "v_proj",
26
  "q_proj"
27
  ],
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b2839b132c4abc0e58aac28c72e2046ef49c371afa54e0ab61553e2f4deb5d2e
3
  size 27297544
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:802d3ab785de55b415eecb965d580c0145fbef3efc693f23754db6d874ef2679
3
  size 27297544
metrics.jsonl CHANGED
@@ -34,3 +34,6 @@
34
  {"epoch": 0.9411764705882353, "precision": 0.0, "recall": 0.0, "fold": 0}
35
  {"epoch": 1.9607843137254903, "precision": 0.0, "recall": 0.0, "fold": 0}
36
  {"epoch": 2.980392156862745, "precision": 0.0, "recall": 0.0, "fold": 0}
 
 
 
 
34
  {"epoch": 0.9411764705882353, "precision": 0.0, "recall": 0.0, "fold": 0}
35
  {"epoch": 1.9607843137254903, "precision": 0.0, "recall": 0.0, "fold": 0}
36
  {"epoch": 2.980392156862745, "precision": 0.0, "recall": 0.0, "fold": 0}
37
+ {"epoch": 0.9411764705882353, "precision": 0.4499999994375, "recall": 0.89999999775, "fold": 0}
38
+ {"epoch": 1.9607843137254903, "precision": 0.5666666657222222, "recall": 0.849999997875, "fold": 0}
39
+ {"epoch": 2.980392156862745, "precision": 0.6734693863806747, "recall": 0.8249999979375, "fold": 0}
metrics_epoch_0.9411764705882353_fold_0_lr_0.0001_seed_9012_weight_5.0.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"epoch": 0.9411764705882353, "precision": 0.4499999994375, "recall": 0.89999999775, "fold": 0}
metrics_epoch_1.9607843137254903_fold_0_lr_0.0001_seed_9012_weight_5.0.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"epoch": 1.9607843137254903, "precision": 0.5666666657222222, "recall": 0.849999997875, "fold": 0}
metrics_epoch_2.980392156862745_fold_0_lr_0.0001_seed_9012_weight_5.0.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"epoch": 2.980392156862745, "precision": 0.6734693863806747, "recall": 0.8249999979375, "fold": 0}
results_epoch_0.9411764705882353_fold_0_lr_0.0001_seed_9012_weight_5.0.json ADDED
The diff for this file is too large to render. See raw diff
 
results_epoch_1.9607843137254903_fold_0_lr_0.0001_seed_9012_weight_5.0.json ADDED
The diff for this file is too large to render. See raw diff
 
results_epoch_2.980392156862745_fold_0_lr_0.0001_seed_9012_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:e383eaf321bb11703b6334e8ba447aa8a47c6a80fc0f1193f8d31661e23eee53
3
  size 5688
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f06a1323ffd1510d415d4befbfce97ba66294e80697d95446a5775786b2e64ea
3
  size 5688