Hugging Face's logo

PaulD
/
llama3_false_positives_1609_KTO_optimised_model

PEFT
Safetensors
trl
kto
Generated from Trainer
Model card Files Community
PaulD commited on
Commit
bb5c1d9
·
verified ·
1 Parent(s): c541858

End of training

Browse files
Files changed (9) hide show
  1. README.md +9 -9
  2. adapter_config.json +1 -1
  3. adapter_model.safetensors +1 -1
  4. metrics.jsonl +2 -2
  5. metrics_epoch_0.96_fold_0_lr_0.0001_seed_42_weight_2.0.json +1 -1
  6. metrics_epoch_2.0_fold_0_lr_0.0001_seed_42_weight_2.0.json +1 -1
  7. results_epoch_0.96_fold_0_lr_0.0001_seed_42_weight_2.0.json +0 -0
  8. results_epoch_2.0_fold_0_lr_0.0001_seed_42_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: 0.5846
22
- - Eval/rewards/chosen: 2.6874
23
- - Eval/logps/chosen: -171.2629
24
- - Eval/rewards/rejected: 2.4373
25
- - Eval/logps/rejected: -186.6407
26
- - Eval/rewards/margins: 0.2501
27
- - Eval/kl: 24.3196
28
 
29
  ## Model description
30
 
@@ -58,8 +58,8 @@ The following hyperparameters were used during training:
58
 
59
  | Training Loss | Epoch | Step | Validation Loss | |
60
  |:-------------:|:-----:|:----:|:---------------:|:-------:|
61
- | 0.4821 | 0.96 | 12 | 0.6001 | 13.3805 |
62
- | 0.1368 | 2.0 | 25 | 0.5846 | 24.3196 |
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: 0.5894
22
+ - Eval/rewards/chosen: 2.7219
23
+ - Eval/logps/chosen: -170.9184
24
+ - Eval/rewards/rejected: 2.5085
25
+ - Eval/logps/rejected: -185.9287
26
+ - Eval/rewards/margins: 0.2134
27
+ - Eval/kl: 25.1066
28
 
29
  ## Model description
30
 
 
58
 
59
  | Training Loss | Epoch | Step | Validation Loss | |
60
  |:-------------:|:-----:|:----:|:---------------:|:-------:|
61
+ | 0.4841 | 0.96 | 12 | 0.6000 | 13.4542 |
62
+ | 0.1395 | 2.0 | 25 | 0.5894 | 25.1066 |
63
 
64
 
65
  ### Framework versions
adapter_config.json CHANGED
@@ -20,9 +20,9 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
23
  "q_proj",
24
  "v_proj",
25
- "k_proj",
26
  "o_proj"
27
  ],
28
  "task_type": "CAUSAL_LM",
 
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",
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:05787831a9c41e4cb5e7a3af9e3e37757bbfad33da93a01f9a36ece2da7e37e2
3
  size 27297544
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb358d3f5971d8aee036a92f94d293ba98b8689725be4ddc1bbafad6bb5db297
3
  size 27297544
metrics.jsonl CHANGED
@@ -1,2 +1,2 @@
1
- {"epoch": 0.96, "precision": 0.6571428552653061, "recall": 0.7666666641111111, "fold": 0}
2
- {"epoch": 2.0, "precision": 0.9999999928571429, "recall": 0.4666666651111111, "fold": 0}
 
1
+ {"epoch": 0.96, "precision": 0.7187499977539062, "recall": 0.7666666641111111, "fold": 0}
2
+ {"epoch": 2.0, "precision": 0.9999999944444444, "recall": 0.599999998, "fold": 0}
metrics_epoch_0.96_fold_0_lr_0.0001_seed_42_weight_2.0.json CHANGED
@@ -1 +1 @@
1
- {"epoch": 0.96, "precision": 0.6571428552653061, "recall": 0.7666666641111111, "fold": 0}
 
1
+ {"epoch": 0.96, "precision": 0.7187499977539062, "recall": 0.7666666641111111, "fold": 0}
metrics_epoch_2.0_fold_0_lr_0.0001_seed_42_weight_2.0.json CHANGED
@@ -1 +1 @@
1
- {"epoch": 2.0, "precision": 0.9999999928571429, "recall": 0.4666666651111111, "fold": 0}
 
1
+ {"epoch": 2.0, "precision": 0.9999999944444444, "recall": 0.599999998, "fold": 0}
results_epoch_0.96_fold_0_lr_0.0001_seed_42_weight_2.0.json CHANGED
The diff for this file is too large to render. See raw diff
 
results_epoch_2.0_fold_0_lr_0.0001_seed_42_weight_2.0.json CHANGED
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:b0fad4ced97704d7d4678501c3db1491a4e4d9db884474ff1d1c3198b578e801
3
  size 5688
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af5dd1598e3302623cc4d6fad11cdfe586c6da34266064325c09caa093724444
3
  size 5688