Hugging Face's logo

aengusl
/
llama2-7b-sft-lora

PEFT
TensorBoard
Safetensors
trl
sft
Generated from Trainer
Model card Files Metrics Community
aengusl commited on
Commit
c9bfaf3
•
1 Parent(s): e003892

Model save

Browse files
Files changed (8) hide show
  1. README.md +6 -6
  2. adapter_config.json +3 -3
  3. runs/Jun11_16-44-25_449b74b344d5/events.out.tfevents.1718124301.449b74b344d5.66196.0 +2 -2
  4. runs/Jun11_16-49-06_449b74b344d5/events.out.tfevents.1718124578.449b74b344d5.70634.0 +3 -0
  5. step_0/README.md +7 -0
  6. step_0/adapter_config.json +3 -3
  7. step_0/training_args.bin +1 -1
  8. training_args.bin +1 -1
README.md CHANGED
@@ -16,7 +16,7 @@ model-index:
16
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
  should probably proofread and complete it, then remove this comment. -->
18
 
19
- [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/quirky_lats_at_mats/harmbench_adv_training/runs/o1b2ladz)
20
  # llama2-7b-sft-lora
21
 
22
  This model is a fine-tuned version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) on the generator dataset.
@@ -38,14 +38,14 @@ More information needed
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
- - learning_rate: 1e-05
42
- - train_batch_size: 4
43
- - eval_batch_size: 8
44
  - seed: 4
45
  - distributed_type: multi-GPU
46
  - num_devices: 2
47
- - total_train_batch_size: 8
48
- - total_eval_batch_size: 16
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: cosine
51
  - training_steps: 2000
 
16
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
  should probably proofread and complete it, then remove this comment. -->
18
 
19
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/quirky_lats_at_mats/harmbench_adv_training/runs/fseyqxvv)
20
  # llama2-7b-sft-lora
21
 
22
  This model is a fine-tuned version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) on the generator dataset.
 
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
+ - learning_rate: 2e-05
42
+ - train_batch_size: 2
43
+ - eval_batch_size: 4
44
  - seed: 4
45
  - distributed_type: multi-GPU
46
  - num_devices: 2
47
+ - total_train_batch_size: 4
48
+ - total_eval_batch_size: 8
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: cosine
51
  - training_steps: 2000
adapter_config.json CHANGED
@@ -17,11 +17,11 @@
17
  "revision": null,
18
  "target_modules": [
19
  "q_proj",
20
- "down_proj",
21
  "k_proj",
22
- "up_proj",
23
  "o_proj",
24
- "v_proj"
 
25
  ],
26
  "task_type": "CAUSAL_LM"
27
  }
 
17
  "revision": null,
18
  "target_modules": [
19
  "q_proj",
20
+ "v_proj",
21
  "k_proj",
 
22
  "o_proj",
23
+ "up_proj",
24
+ "down_proj"
25
  ],
26
  "task_type": "CAUSAL_LM"
27
  }
runs/Jun11_16-44-25_449b74b344d5/events.out.tfevents.1718124301.449b74b344d5.66196.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:359a46c3dc7bd67428e3b2aed8f0d20c5d2bf931b8a86afbba81ba8a6e4481ff
3
- size 5069
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:765163c76e58799566f79bcd95c3a0a2c3d389a818fb37517c987ab9975e2f7b
3
+ size 5377
runs/Jun11_16-49-06_449b74b344d5/events.out.tfevents.1718124578.449b74b344d5.70634.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0346ed6e41c1f779ccd1a107cac8d1ce52425c126fc3c1bcacb5bca42aa186b1
3
+ size 5069
step_0/README.md CHANGED
@@ -665,4 +665,11 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
665
  ### Framework versions
666
 
667
 
 
 
 
 
 
 
 
668
  - PEFT 0.6.1
 
665
  ### Framework versions
666
 
667
 
668
+ - PEFT 0.6.1
669
+ ## Training procedure
670
+
671
+
672
+ ### Framework versions
673
+
674
+
675
  - PEFT 0.6.1
step_0/adapter_config.json CHANGED
@@ -17,11 +17,11 @@
17
  "revision": null,
18
  "target_modules": [
19
  "q_proj",
20
- "down_proj",
21
  "k_proj",
22
- "up_proj",
23
  "o_proj",
24
- "v_proj"
 
25
  ],
26
  "task_type": "CAUSAL_LM"
27
  }
 
17
  "revision": null,
18
  "target_modules": [
19
  "q_proj",
20
+ "v_proj",
21
  "k_proj",
 
22
  "o_proj",
23
+ "up_proj",
24
+ "down_proj"
25
  ],
26
  "task_type": "CAUSAL_LM"
27
  }
step_0/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d6edec78807b180c3fb614d22e18f16d4aa34fbadf52a968c0bd3884a85e1de2
3
  size 6200
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ff25ef904e88b2406ea09463d847795f149c9d0bf1d5c2ffd75b61d57cbf327
3
  size 6200
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d6edec78807b180c3fb614d22e18f16d4aa34fbadf52a968c0bd3884a85e1de2
3
  size 6200
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ff25ef904e88b2406ea09463d847795f149c9d0bf1d5c2ffd75b61d57cbf327
3
  size 6200