Hugging Face's logo

wangrongsheng
/
LLM

Model card Files
xet
Community
wangrongsheng commited on
Commit
42de9a6
·
1 Parent(s): 65133da

commit from root

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. ppo-lora/README.md +3 -0
  2. ppo-lora/adapter_config.json +20 -0
  3. ppo-lora/adapter_model.bin +3 -0
  4. ppo-lora/checkpoint-1000/README.md +3 -0
  5. ppo-lora/checkpoint-1000/adapter_config.json +20 -0
  6. ppo-lora/checkpoint-1000/adapter_model.bin +3 -0
  7. ppo-lora/checkpoint-1000/finetuning_args.json +13 -0
  8. ppo-lora/checkpoint-1000/reward/adapter_config.json +20 -0
  9. ppo-lora/checkpoint-1000/reward/adapter_model.bin +3 -0
  10. ppo-lora/checkpoint-1000/training_args.bin +3 -0
  11. ppo-lora/checkpoint-1000/value_head.bin +3 -0
  12. ppo-lora/checkpoint-2000/README.md +3 -0
  13. ppo-lora/checkpoint-2000/adapter_config.json +20 -0
  14. ppo-lora/checkpoint-2000/adapter_model.bin +3 -0
  15. ppo-lora/checkpoint-2000/finetuning_args.json +13 -0
  16. ppo-lora/checkpoint-2000/reward/adapter_config.json +20 -0
  17. ppo-lora/checkpoint-2000/reward/adapter_model.bin +3 -0
  18. ppo-lora/checkpoint-2000/training_args.bin +3 -0
  19. ppo-lora/checkpoint-2000/value_head.bin +3 -0
  20. ppo-lora/checkpoint-3000/README.md +3 -0
  21. ppo-lora/checkpoint-3000/adapter_config.json +20 -0
  22. ppo-lora/checkpoint-3000/adapter_model.bin +3 -0
  23. ppo-lora/checkpoint-3000/finetuning_args.json +13 -0
  24. ppo-lora/checkpoint-3000/reward/adapter_config.json +20 -0
  25. ppo-lora/checkpoint-3000/reward/adapter_model.bin +3 -0
  26. ppo-lora/checkpoint-3000/training_args.bin +3 -0
  27. ppo-lora/checkpoint-3000/value_head.bin +3 -0
  28. ppo-lora/checkpoint-4000/README.md +3 -0
  29. ppo-lora/checkpoint-4000/adapter_config.json +20 -0
  30. ppo-lora/checkpoint-4000/adapter_model.bin +3 -0
  31. ppo-lora/checkpoint-4000/finetuning_args.json +13 -0
  32. ppo-lora/checkpoint-4000/reward/adapter_config.json +20 -0
  33. ppo-lora/checkpoint-4000/reward/adapter_model.bin +3 -0
  34. ppo-lora/checkpoint-4000/training_args.bin +3 -0
  35. ppo-lora/checkpoint-4000/value_head.bin +3 -0
  36. ppo-lora/checkpoint-5000/README.md +3 -0
  37. ppo-lora/checkpoint-5000/adapter_config.json +20 -0
  38. ppo-lora/checkpoint-5000/adapter_model.bin +3 -0
  39. ppo-lora/checkpoint-5000/finetuning_args.json +13 -0
  40. ppo-lora/checkpoint-5000/reward/adapter_config.json +20 -0
  41. ppo-lora/checkpoint-5000/reward/adapter_model.bin +3 -0
  42. ppo-lora/checkpoint-5000/training_args.bin +3 -0
  43. ppo-lora/checkpoint-5000/value_head.bin +3 -0
  44. ppo-lora/checkpoint-6000/README.md +3 -0
  45. ppo-lora/checkpoint-6000/adapter_config.json +20 -0
  46. ppo-lora/checkpoint-6000/adapter_model.bin +3 -0
  47. ppo-lora/checkpoint-6000/finetuning_args.json +13 -0
  48. ppo-lora/checkpoint-6000/reward/adapter_config.json +20 -0
  49. ppo-lora/checkpoint-6000/reward/adapter_model.bin +3 -0
  50. ppo-lora/checkpoint-6000/training_args.bin +3 -0
ppo-lora/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
ppo-lora/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55c6ec1b08f5cea38e43c3e4eb11b0be787f3cd86a46cf09fe0e6e5048737dea
3
+ size 26269517
ppo-lora/checkpoint-1000/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
ppo-lora/checkpoint-1000/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-1000/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c213d0e4f8d874939ab229e712e25c78ddc973176bd844440186e4b8977c4651
3
+ size 26269517
ppo-lora/checkpoint-1000/finetuning_args.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "finetuning_type": "lora",
3
+ "lora_alpha": 32.0,
4
+ "lora_dropout": 0.1,
5
+ "lora_rank": 8,
6
+ "lora_target": [
7
+ "q_proj",
8
+ "v_proj"
9
+ ],
10
+ "name_module_trainable": "mlp",
11
+ "num_hidden_layers": 32,
12
+ "num_layer_trainable": 3
13
+ }
ppo-lora/checkpoint-1000/reward/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-1000/reward/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
3
+ size 443
ppo-lora/checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf91d7ac2ddda7e1272b902408951545188f843d4da6b19c2e572a6c86a6075b
3
+ size 3274
ppo-lora/checkpoint-1000/value_head.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1da5b73c9aa13a16f0875a3c73b3ec6df71a08d94ce82197b9a3d2f077f30a95
3
+ size 21491
ppo-lora/checkpoint-2000/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
ppo-lora/checkpoint-2000/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-2000/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d0fbeb1c6409498c16a2c2f2e70b7446bf4ac1234fb8dc2f8df9806b2462ddb
3
+ size 26269517
ppo-lora/checkpoint-2000/finetuning_args.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "finetuning_type": "lora",
3
+ "lora_alpha": 32.0,
4
+ "lora_dropout": 0.1,
5
+ "lora_rank": 8,
6
+ "lora_target": [
7
+ "q_proj",
8
+ "v_proj"
9
+ ],
10
+ "name_module_trainable": "mlp",
11
+ "num_hidden_layers": 32,
12
+ "num_layer_trainable": 3
13
+ }
ppo-lora/checkpoint-2000/reward/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-2000/reward/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
3
+ size 443
ppo-lora/checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf91d7ac2ddda7e1272b902408951545188f843d4da6b19c2e572a6c86a6075b
3
+ size 3274
ppo-lora/checkpoint-2000/value_head.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9c26a0f0760505ea0f2efdc4bd9d8f0ceb4bd4a0d23211e24d96b907f4c0028
3
+ size 21491
ppo-lora/checkpoint-3000/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
ppo-lora/checkpoint-3000/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-3000/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ad9d239c7787f8d73de9875ae02a52464859764422d88a4970e344b4fd9faee
3
+ size 26269517
ppo-lora/checkpoint-3000/finetuning_args.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "finetuning_type": "lora",
3
+ "lora_alpha": 32.0,
4
+ "lora_dropout": 0.1,
5
+ "lora_rank": 8,
6
+ "lora_target": [
7
+ "q_proj",
8
+ "v_proj"
9
+ ],
10
+ "name_module_trainable": "mlp",
11
+ "num_hidden_layers": 32,
12
+ "num_layer_trainable": 3
13
+ }
ppo-lora/checkpoint-3000/reward/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-3000/reward/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
3
+ size 443
ppo-lora/checkpoint-3000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf91d7ac2ddda7e1272b902408951545188f843d4da6b19c2e572a6c86a6075b
3
+ size 3274
ppo-lora/checkpoint-3000/value_head.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3817aaf3884ab53bafa9a0a901eb2cc3329b55238d28a98b937ce5db5e888a9
3
+ size 21491
ppo-lora/checkpoint-4000/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
ppo-lora/checkpoint-4000/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-4000/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e11cb62dcf8c47decf25bb4525c4921aa48aee0815306027ec7d96a32a459ae
3
+ size 26269517
ppo-lora/checkpoint-4000/finetuning_args.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "finetuning_type": "lora",
3
+ "lora_alpha": 32.0,
4
+ "lora_dropout": 0.1,
5
+ "lora_rank": 8,
6
+ "lora_target": [
7
+ "q_proj",
8
+ "v_proj"
9
+ ],
10
+ "name_module_trainable": "mlp",
11
+ "num_hidden_layers": 32,
12
+ "num_layer_trainable": 3
13
+ }
ppo-lora/checkpoint-4000/reward/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-4000/reward/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
3
+ size 443
ppo-lora/checkpoint-4000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf91d7ac2ddda7e1272b902408951545188f843d4da6b19c2e572a6c86a6075b
3
+ size 3274
ppo-lora/checkpoint-4000/value_head.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fca5a9037d28fec50bc06f7b4768851167c5dbac08df3d6f579c0fc86e67c530
3
+ size 21491
ppo-lora/checkpoint-5000/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
ppo-lora/checkpoint-5000/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-5000/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06839ee491e98e367765e336e05ec2ea6133406f411a33ae8b054c77dddb1513
3
+ size 26269517
ppo-lora/checkpoint-5000/finetuning_args.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "finetuning_type": "lora",
3
+ "lora_alpha": 32.0,
4
+ "lora_dropout": 0.1,
5
+ "lora_rank": 8,
6
+ "lora_target": [
7
+ "q_proj",
8
+ "v_proj"
9
+ ],
10
+ "name_module_trainable": "mlp",
11
+ "num_hidden_layers": 32,
12
+ "num_layer_trainable": 3
13
+ }
ppo-lora/checkpoint-5000/reward/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-5000/reward/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
3
+ size 443
ppo-lora/checkpoint-5000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf91d7ac2ddda7e1272b902408951545188f843d4da6b19c2e572a6c86a6075b
3
+ size 3274
ppo-lora/checkpoint-5000/value_head.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:404a4c61cdfa78d1d30457f1956181a78bd80eb3396e99ef0ed0cc40570d3ea3
3
+ size 21491
ppo-lora/checkpoint-6000/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
ppo-lora/checkpoint-6000/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-6000/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0424271468f1a8d78da73c08a499c2e6a19aa5df3104bb1f83d462e37d39300e
3
+ size 26269517
ppo-lora/checkpoint-6000/finetuning_args.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "finetuning_type": "lora",
3
+ "lora_alpha": 32.0,
4
+ "lora_dropout": 0.1,
5
+ "lora_rank": 8,
6
+ "lora_target": [
7
+ "q_proj",
8
+ "v_proj"
9
+ ],
10
+ "name_module_trainable": "mlp",
11
+ "num_hidden_layers": 32,
12
+ "num_layer_trainable": 3
13
+ }
ppo-lora/checkpoint-6000/reward/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "llama-13b-hf/",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 32.0,
10
+ "lora_dropout": 0.1,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
ppo-lora/checkpoint-6000/reward/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
3
+ size 443
ppo-lora/checkpoint-6000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf91d7ac2ddda7e1272b902408951545188f843d4da6b19c2e572a6c86a6075b
3
+ size 3274