Initial LoRA finetuned model and tokenizer for CoT
Browse files- adapter_config.json +4 -4
- adapter_model.safetensors +2 -2
- added_tokens.json +3 -0
- checkpoint-7/adapter_config.json +4 -4
- checkpoint-7/adapter_model.safetensors +2 -2
- checkpoint-7/added_tokens.json +3 -0
- checkpoint-7/optimizer.pt +1 -1
- checkpoint-7/rng_state.pth +1 -1
- checkpoint-7/special_tokens_map.json +7 -1
- checkpoint-7/tokenizer.json +9 -0
- checkpoint-7/tokenizer_config.json +9 -1
- checkpoint-7/trainer_state.json +1 -1
- checkpoint-7/training_args.bin +1 -1
- special_tokens_map.json +7 -1
- tokenizer.json +9 -0
- tokenizer_config.json +9 -1
adapter_config.json
CHANGED
|
@@ -25,12 +25,12 @@
|
|
| 25 |
"revision": null,
|
| 26 |
"target_modules": [
|
| 27 |
"down_proj",
|
| 28 |
-
"q_proj",
|
| 29 |
-
"up_proj",
|
| 30 |
-
"gate_proj",
|
| 31 |
"k_proj",
|
|
|
|
| 32 |
"o_proj",
|
| 33 |
-
"
|
|
|
|
|
|
|
| 34 |
],
|
| 35 |
"task_type": "CAUSAL_LM",
|
| 36 |
"trainable_token_indices": null,
|
|
|
|
| 25 |
"revision": null,
|
| 26 |
"target_modules": [
|
| 27 |
"down_proj",
|
|
|
|
|
|
|
|
|
|
| 28 |
"k_proj",
|
| 29 |
+
"v_proj",
|
| 30 |
"o_proj",
|
| 31 |
+
"gate_proj",
|
| 32 |
+
"up_proj",
|
| 33 |
+
"q_proj"
|
| 34 |
],
|
| 35 |
"task_type": "CAUSAL_LM",
|
| 36 |
"trainable_token_indices": null,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b285968d22cd626569dfe97dea93f0cb395118921d3f63f125ede9ba83b046d9
|
| 3 |
+
size 1986736152
|
added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[PAD]": 43176
|
| 3 |
+
}
|
checkpoint-7/adapter_config.json
CHANGED
|
@@ -25,12 +25,12 @@
|
|
| 25 |
"revision": null,
|
| 26 |
"target_modules": [
|
| 27 |
"down_proj",
|
| 28 |
-
"q_proj",
|
| 29 |
-
"up_proj",
|
| 30 |
-
"gate_proj",
|
| 31 |
"k_proj",
|
|
|
|
| 32 |
"o_proj",
|
| 33 |
-
"
|
|
|
|
|
|
|
| 34 |
],
|
| 35 |
"task_type": "CAUSAL_LM",
|
| 36 |
"trainable_token_indices": null,
|
|
|
|
| 25 |
"revision": null,
|
| 26 |
"target_modules": [
|
| 27 |
"down_proj",
|
|
|
|
|
|
|
|
|
|
| 28 |
"k_proj",
|
| 29 |
+
"v_proj",
|
| 30 |
"o_proj",
|
| 31 |
+
"gate_proj",
|
| 32 |
+
"up_proj",
|
| 33 |
+
"q_proj"
|
| 34 |
],
|
| 35 |
"task_type": "CAUSAL_LM",
|
| 36 |
"trainable_token_indices": null,
|
checkpoint-7/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b285968d22cd626569dfe97dea93f0cb395118921d3f63f125ede9ba83b046d9
|
| 3 |
+
size 1986736152
|
checkpoint-7/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[PAD]": 43176
|
| 3 |
+
}
|
checkpoint-7/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 650163476
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a0161e9e9e182ee383d1226d219fb51eee7a0a7dee2d3cab2c3ef156e602996f
|
| 3 |
size 650163476
|
checkpoint-7/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:63c72f5771f7347224a4a09ed2354c66aedcc180caa4df8fc886e7d8acb9a307
|
| 3 |
size 14244
|
checkpoint-7/special_tokens_map.json
CHANGED
|
@@ -13,7 +13,13 @@
|
|
| 13 |
"rstrip": false,
|
| 14 |
"single_word": false
|
| 15 |
},
|
| 16 |
-
"pad_token":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
"unk_token": {
|
| 18 |
"content": "<unk>",
|
| 19 |
"lstrip": false,
|
|
|
|
| 13 |
"rstrip": false,
|
| 14 |
"single_word": false
|
| 15 |
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
"unk_token": {
|
| 24 |
"content": "<unk>",
|
| 25 |
"lstrip": false,
|
checkpoint-7/tokenizer.json
CHANGED
|
@@ -29,6 +29,15 @@
|
|
| 29 |
"rstrip": false,
|
| 30 |
"normalized": true,
|
| 31 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
}
|
| 33 |
],
|
| 34 |
"normalizer": null,
|
|
|
|
| 29 |
"rstrip": false,
|
| 30 |
"normalized": true,
|
| 31 |
"special": true
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"id": 43176,
|
| 35 |
+
"content": "[PAD]",
|
| 36 |
+
"single_word": false,
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"rstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"special": true
|
| 41 |
}
|
| 42 |
],
|
| 43 |
"normalizer": null,
|
checkpoint-7/tokenizer_config.json
CHANGED
|
@@ -26,6 +26,14 @@
|
|
| 26 |
"rstrip": false,
|
| 27 |
"single_word": false,
|
| 28 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
}
|
| 30 |
},
|
| 31 |
"bos_token": "<s>",
|
|
@@ -34,7 +42,7 @@
|
|
| 34 |
"extra_special_tokens": {},
|
| 35 |
"legacy": false,
|
| 36 |
"model_max_length": 1000000000000000019884624838656,
|
| 37 |
-
"pad_token": "
|
| 38 |
"padding_side": "right",
|
| 39 |
"sp_model_kwargs": {},
|
| 40 |
"spaces_between_special_tokens": false,
|
|
|
|
| 26 |
"rstrip": false,
|
| 27 |
"single_word": false,
|
| 28 |
"special": true
|
| 29 |
+
},
|
| 30 |
+
"43176": {
|
| 31 |
+
"content": "[PAD]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
}
|
| 38 |
},
|
| 39 |
"bos_token": "<s>",
|
|
|
|
| 42 |
"extra_special_tokens": {},
|
| 43 |
"legacy": false,
|
| 44 |
"model_max_length": 1000000000000000019884624838656,
|
| 45 |
+
"pad_token": "[PAD]",
|
| 46 |
"padding_side": "right",
|
| 47 |
"sp_model_kwargs": {},
|
| 48 |
"spaces_between_special_tokens": false,
|
checkpoint-7/trainer_state.json
CHANGED
|
@@ -26,7 +26,7 @@
|
|
| 26 |
"attributes": {}
|
| 27 |
}
|
| 28 |
},
|
| 29 |
-
"total_flos":
|
| 30 |
"train_batch_size": 2,
|
| 31 |
"trial_name": null,
|
| 32 |
"trial_params": null
|
|
|
|
| 26 |
"attributes": {}
|
| 27 |
}
|
| 28 |
},
|
| 29 |
+
"total_flos": 4284174512947200.0,
|
| 30 |
"train_batch_size": 2,
|
| 31 |
"trial_name": null,
|
| 32 |
"trial_params": null
|
checkpoint-7/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5368
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:35872fcc6d491acc7effa1c32a80c35245f0508a790d9d56ac5693a9faca89eb
|
| 3 |
size 5368
|
special_tokens_map.json
CHANGED
|
@@ -13,7 +13,13 @@
|
|
| 13 |
"rstrip": false,
|
| 14 |
"single_word": false
|
| 15 |
},
|
| 16 |
-
"pad_token":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
"unk_token": {
|
| 18 |
"content": "<unk>",
|
| 19 |
"lstrip": false,
|
|
|
|
| 13 |
"rstrip": false,
|
| 14 |
"single_word": false
|
| 15 |
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
"unk_token": {
|
| 24 |
"content": "<unk>",
|
| 25 |
"lstrip": false,
|
tokenizer.json
CHANGED
|
@@ -29,6 +29,15 @@
|
|
| 29 |
"rstrip": false,
|
| 30 |
"normalized": true,
|
| 31 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
}
|
| 33 |
],
|
| 34 |
"normalizer": null,
|
|
|
|
| 29 |
"rstrip": false,
|
| 30 |
"normalized": true,
|
| 31 |
"special": true
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"id": 43176,
|
| 35 |
+
"content": "[PAD]",
|
| 36 |
+
"single_word": false,
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"rstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"special": true
|
| 41 |
}
|
| 42 |
],
|
| 43 |
"normalizer": null,
|
tokenizer_config.json
CHANGED
|
@@ -26,6 +26,14 @@
|
|
| 26 |
"rstrip": false,
|
| 27 |
"single_word": false,
|
| 28 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
}
|
| 30 |
},
|
| 31 |
"bos_token": "<s>",
|
|
@@ -34,7 +42,7 @@
|
|
| 34 |
"extra_special_tokens": {},
|
| 35 |
"legacy": false,
|
| 36 |
"model_max_length": 1000000000000000019884624838656,
|
| 37 |
-
"pad_token": "
|
| 38 |
"padding_side": "right",
|
| 39 |
"sp_model_kwargs": {},
|
| 40 |
"spaces_between_special_tokens": false,
|
|
|
|
| 26 |
"rstrip": false,
|
| 27 |
"single_word": false,
|
| 28 |
"special": true
|
| 29 |
+
},
|
| 30 |
+
"43176": {
|
| 31 |
+
"content": "[PAD]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
}
|
| 38 |
},
|
| 39 |
"bos_token": "<s>",
|
|
|
|
| 42 |
"extra_special_tokens": {},
|
| 43 |
"legacy": false,
|
| 44 |
"model_max_length": 1000000000000000019884624838656,
|
| 45 |
+
"pad_token": "[PAD]",
|
| 46 |
"padding_side": "right",
|
| 47 |
"sp_model_kwargs": {},
|
| 48 |
"spaces_between_special_tokens": false,
|