1.5
Browse files- README.md +83 -117
- adapter_config.json +5 -5
- adapter_model.safetensors +1 -1
- all_results.json +10 -10
- eval_results.json +5 -5
- train_results.json +6 -6
- trainer_log.jsonl +0 -0
- trainer_state.json +0 -0
- training_args.bin +1 -1
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -1,25 +1,25 @@
|
|
| 1 |
---
|
| 2 |
library_name: peft
|
| 3 |
license: other
|
| 4 |
-
base_model: unsloth/
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
- lora
|
| 8 |
- unsloth
|
| 9 |
- generated_from_trainer
|
| 10 |
model-index:
|
| 11 |
-
- name: llm3br256
|
| 12 |
results: []
|
| 13 |
---
|
| 14 |
|
| 15 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
|
| 18 |
-
# llm3br256
|
| 19 |
|
| 20 |
This model is a fine-tuned version of [meta-llama/Llama-3.2-3B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct) on the asianpaints dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
-
- Loss: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
@@ -39,128 +39,94 @@ More information needed
|
|
| 39 |
|
| 40 |
The following hyperparameters were used during training:
|
| 41 |
- learning_rate: 0.0001
|
| 42 |
-
- train_batch_size:
|
| 43 |
-
- eval_batch_size:
|
| 44 |
- seed: 42
|
| 45 |
-
- gradient_accumulation_steps: 8
|
| 46 |
-
- total_train_batch_size: 32
|
| 47 |
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 48 |
- lr_scheduler_type: cosine
|
| 49 |
- lr_scheduler_warmup_ratio: 0.1
|
| 50 |
-
- num_epochs:
|
| 51 |
|
| 52 |
### Training results
|
| 53 |
|
| 54 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
|:-------------:|:------:|:----:|:---------------:|
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.
|
| 76 |
-
| 0.
|
| 77 |
-
| 0.
|
| 78 |
-
| 0.
|
| 79 |
-
| 0.
|
| 80 |
-
| 0.
|
| 81 |
-
| 0.
|
| 82 |
-
| 0.
|
| 83 |
-
| 0.
|
| 84 |
-
| 0.
|
| 85 |
-
| 0.
|
| 86 |
-
| 0.
|
| 87 |
-
| 0.
|
| 88 |
-
| 0.
|
| 89 |
-
| 0.
|
| 90 |
-
| 0.
|
| 91 |
-
| 0.
|
| 92 |
-
| 0.
|
| 93 |
-
| 0.
|
| 94 |
-
| 0.
|
| 95 |
-
| 0.0115 |
|
| 96 |
-
| 0.
|
| 97 |
-
| 0.
|
| 98 |
-
| 0.
|
| 99 |
-
| 0.
|
| 100 |
-
| 0.
|
| 101 |
-
| 0.
|
| 102 |
-
| 0.
|
| 103 |
-
| 0.
|
| 104 |
-
| 0.
|
| 105 |
-
| 0.
|
| 106 |
-
| 0.
|
| 107 |
-
| 0.
|
| 108 |
-
| 0.
|
| 109 |
-
| 0.
|
| 110 |
-
| 0.
|
| 111 |
-
| 0.
|
| 112 |
-
| 0.
|
| 113 |
-
| 0.
|
| 114 |
-
| 0.
|
| 115 |
-
| 0.
|
| 116 |
-
| 0.
|
| 117 |
-
| 0.
|
| 118 |
-
| 0.
|
| 119 |
-
| 0.
|
| 120 |
-
| 0.
|
| 121 |
-
| 0.
|
| 122 |
-
| 0.
|
| 123 |
-
| 0.
|
| 124 |
-
| 0.
|
| 125 |
-
| 0.
|
| 126 |
-
| 0.
|
| 127 |
-
| 0.
|
| 128 |
-
| 0.
|
| 129 |
-
| 0.
|
| 130 |
-
| 0.
|
| 131 |
-
| 0.
|
| 132 |
-
| 0.0076 | 3.5402 | 385 | 0.0116 |
|
| 133 |
-
| 0.0068 | 3.5862 | 390 | 0.0114 |
|
| 134 |
-
| 0.0074 | 3.6322 | 395 | 0.0115 |
|
| 135 |
-
| 0.0065 | 3.6782 | 400 | 0.0114 |
|
| 136 |
-
| 0.007 | 3.7241 | 405 | 0.0112 |
|
| 137 |
-
| 0.0064 | 3.7701 | 410 | 0.0112 |
|
| 138 |
-
| 0.0073 | 3.8161 | 415 | 0.0111 |
|
| 139 |
-
| 0.0065 | 3.8621 | 420 | 0.0113 |
|
| 140 |
-
| 0.0069 | 3.9080 | 425 | 0.0111 |
|
| 141 |
-
| 0.0065 | 3.9540 | 430 | 0.0111 |
|
| 142 |
-
| 0.0076 | 4.0 | 435 | 0.0111 |
|
| 143 |
-
| 0.0047 | 4.0460 | 440 | 0.0115 |
|
| 144 |
-
| 0.0053 | 4.0920 | 445 | 0.0119 |
|
| 145 |
-
| 0.0053 | 4.1379 | 450 | 0.0120 |
|
| 146 |
-
| 0.0055 | 4.1839 | 455 | 0.0119 |
|
| 147 |
-
| 0.0053 | 4.2299 | 460 | 0.0117 |
|
| 148 |
-
| 0.0053 | 4.2759 | 465 | 0.0117 |
|
| 149 |
-
| 0.0053 | 4.3218 | 470 | 0.0117 |
|
| 150 |
-
| 0.0058 | 4.3678 | 475 | 0.0116 |
|
| 151 |
-
| 0.0053 | 4.4138 | 480 | 0.0116 |
|
| 152 |
-
| 0.0053 | 4.4598 | 485 | 0.0118 |
|
| 153 |
-
| 0.0051 | 4.5057 | 490 | 0.0117 |
|
| 154 |
-
| 0.0053 | 4.5517 | 495 | 0.0117 |
|
| 155 |
-
| 0.0059 | 4.5977 | 500 | 0.0117 |
|
| 156 |
-
| 0.0055 | 4.6437 | 505 | 0.0117 |
|
| 157 |
-
| 0.0054 | 4.6897 | 510 | 0.0116 |
|
| 158 |
-
| 0.0055 | 4.7356 | 515 | 0.0117 |
|
| 159 |
-
| 0.0056 | 4.7816 | 520 | 0.0116 |
|
| 160 |
-
| 0.0048 | 4.8276 | 525 | 0.0116 |
|
| 161 |
-
| 0.0049 | 4.8736 | 530 | 0.0116 |
|
| 162 |
-
| 0.0043 | 4.9195 | 535 | 0.0116 |
|
| 163 |
-
| 0.0046 | 4.9655 | 540 | 0.0116 |
|
| 164 |
|
| 165 |
|
| 166 |
### Framework versions
|
|
|
|
| 1 |
---
|
| 2 |
library_name: peft
|
| 3 |
license: other
|
| 4 |
+
base_model: unsloth/llama-3.2-3b-instruct-bnb-4bit
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
- lora
|
| 8 |
- unsloth
|
| 9 |
- generated_from_trainer
|
| 10 |
model-index:
|
| 11 |
+
- name: llm3br256-v1.5
|
| 12 |
results: []
|
| 13 |
---
|
| 14 |
|
| 15 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
|
| 18 |
+
# llm3br256-v1.5
|
| 19 |
|
| 20 |
This model is a fine-tuned version of [meta-llama/Llama-3.2-3B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct) on the asianpaints dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.0157
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 39 |
|
| 40 |
The following hyperparameters were used during training:
|
| 41 |
- learning_rate: 0.0001
|
| 42 |
+
- train_batch_size: 48
|
| 43 |
+
- eval_batch_size: 48
|
| 44 |
- seed: 42
|
|
|
|
|
|
|
| 45 |
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 46 |
- lr_scheduler_type: cosine
|
| 47 |
- lr_scheduler_warmup_ratio: 0.1
|
| 48 |
+
- num_epochs: 25
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
|:-------------:|:------:|:----:|:---------------:|
|
| 54 |
+
| 0.1302 | 0.1208 | 25 | 0.1429 |
|
| 55 |
+
| 0.0909 | 0.2415 | 50 | 0.0968 |
|
| 56 |
+
| 0.062 | 0.3623 | 75 | 0.0748 |
|
| 57 |
+
| 0.0602 | 0.4831 | 100 | 0.0608 |
|
| 58 |
+
| 0.057 | 0.6039 | 125 | 0.0552 |
|
| 59 |
+
| 0.0456 | 0.7246 | 150 | 0.0501 |
|
| 60 |
+
| 0.0432 | 0.8454 | 175 | 0.0470 |
|
| 61 |
+
| 0.0416 | 0.9662 | 200 | 0.0447 |
|
| 62 |
+
| 0.0441 | 1.0870 | 225 | 0.0428 |
|
| 63 |
+
| 0.0348 | 1.2077 | 250 | 0.0405 |
|
| 64 |
+
| 0.0355 | 1.3285 | 275 | 0.0386 |
|
| 65 |
+
| 0.0379 | 1.4493 | 300 | 0.0358 |
|
| 66 |
+
| 0.032 | 1.5700 | 325 | 0.0354 |
|
| 67 |
+
| 0.0342 | 1.6908 | 350 | 0.0335 |
|
| 68 |
+
| 0.0318 | 1.8116 | 375 | 0.0324 |
|
| 69 |
+
| 0.031 | 1.9324 | 400 | 0.0318 |
|
| 70 |
+
| 0.0283 | 2.0531 | 425 | 0.0321 |
|
| 71 |
+
| 0.0275 | 2.1739 | 450 | 0.0337 |
|
| 72 |
+
| 0.026 | 2.2947 | 475 | 0.0314 |
|
| 73 |
+
| 0.0244 | 2.4155 | 500 | 0.0285 |
|
| 74 |
+
| 0.0281 | 2.5362 | 525 | 0.0285 |
|
| 75 |
+
| 0.0212 | 2.6570 | 550 | 0.0268 |
|
| 76 |
+
| 0.0221 | 2.7778 | 575 | 0.0267 |
|
| 77 |
+
| 0.0225 | 2.8986 | 600 | 0.0266 |
|
| 78 |
+
| 0.0264 | 3.0193 | 625 | 0.0292 |
|
| 79 |
+
| 0.0196 | 3.1401 | 650 | 0.0280 |
|
| 80 |
+
| 0.0185 | 3.2609 | 675 | 0.0264 |
|
| 81 |
+
| 0.0161 | 3.3816 | 700 | 0.0248 |
|
| 82 |
+
| 0.0186 | 3.5024 | 725 | 0.0226 |
|
| 83 |
+
| 0.0166 | 3.6232 | 750 | 0.0213 |
|
| 84 |
+
| 0.0141 | 3.7440 | 775 | 0.0215 |
|
| 85 |
+
| 0.0186 | 3.8647 | 800 | 0.0211 |
|
| 86 |
+
| 0.0119 | 3.9855 | 825 | 0.0204 |
|
| 87 |
+
| 0.0097 | 4.1063 | 850 | 0.0210 |
|
| 88 |
+
| 0.0095 | 4.2271 | 875 | 0.0204 |
|
| 89 |
+
| 0.0119 | 4.3478 | 900 | 0.0207 |
|
| 90 |
+
| 0.0131 | 4.4686 | 925 | 0.0257 |
|
| 91 |
+
| 0.0123 | 4.5894 | 950 | 0.0228 |
|
| 92 |
+
| 0.0133 | 4.7101 | 975 | 0.0204 |
|
| 93 |
+
| 0.0115 | 4.8309 | 1000 | 0.0191 |
|
| 94 |
+
| 0.0152 | 4.9517 | 1025 | 0.0201 |
|
| 95 |
+
| 0.0075 | 5.0725 | 1050 | 0.0188 |
|
| 96 |
+
| 0.0069 | 5.1932 | 1075 | 0.0169 |
|
| 97 |
+
| 0.0073 | 5.3140 | 1100 | 0.0182 |
|
| 98 |
+
| 0.0076 | 5.4348 | 1125 | 0.0166 |
|
| 99 |
+
| 0.0084 | 5.5556 | 1150 | 0.0173 |
|
| 100 |
+
| 0.0091 | 5.6763 | 1175 | 0.0175 |
|
| 101 |
+
| 0.0081 | 5.7971 | 1200 | 0.0176 |
|
| 102 |
+
| 0.0071 | 5.9179 | 1225 | 0.0175 |
|
| 103 |
+
| 0.0058 | 6.0386 | 1250 | 0.0187 |
|
| 104 |
+
| 0.0081 | 6.1594 | 1275 | 0.0165 |
|
| 105 |
+
| 0.0057 | 6.2802 | 1300 | 0.0171 |
|
| 106 |
+
| 0.0068 | 6.4010 | 1325 | 0.0165 |
|
| 107 |
+
| 0.0059 | 6.5217 | 1350 | 0.0163 |
|
| 108 |
+
| 0.0057 | 6.6425 | 1375 | 0.0151 |
|
| 109 |
+
| 0.0061 | 6.7633 | 1400 | 0.0164 |
|
| 110 |
+
| 0.006 | 6.8841 | 1425 | 0.0156 |
|
| 111 |
+
| 0.0062 | 7.0048 | 1450 | 0.0161 |
|
| 112 |
+
| 0.006 | 7.1256 | 1475 | 0.0178 |
|
| 113 |
+
| 0.0059 | 7.2464 | 1500 | 0.0169 |
|
| 114 |
+
| 0.0043 | 7.3671 | 1525 | 0.0175 |
|
| 115 |
+
| 0.0049 | 7.4879 | 1550 | 0.0178 |
|
| 116 |
+
| 0.0058 | 7.6087 | 1575 | 0.0156 |
|
| 117 |
+
| 0.0062 | 7.7295 | 1600 | 0.0158 |
|
| 118 |
+
| 0.0045 | 7.8502 | 1625 | 0.0151 |
|
| 119 |
+
| 0.0054 | 7.9710 | 1650 | 0.0150 |
|
| 120 |
+
| 0.0042 | 8.0918 | 1675 | 0.0157 |
|
| 121 |
+
| 0.0039 | 8.2126 | 1700 | 0.0157 |
|
| 122 |
+
| 0.0046 | 8.3333 | 1725 | 0.0170 |
|
| 123 |
+
| 0.0025 | 8.4541 | 1750 | 0.0154 |
|
| 124 |
+
| 0.0047 | 8.5749 | 1775 | 0.0156 |
|
| 125 |
+
| 0.0044 | 8.6957 | 1800 | 0.0166 |
|
| 126 |
+
| 0.0031 | 8.8164 | 1825 | 0.0172 |
|
| 127 |
+
| 0.0029 | 8.9372 | 1850 | 0.0167 |
|
| 128 |
+
| 0.0032 | 9.0580 | 1875 | 0.0169 |
|
| 129 |
+
| 0.0036 | 9.1787 | 1900 | 0.0167 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 130 |
|
| 131 |
|
| 132 |
### Framework versions
|
adapter_config.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"alpha_pattern": {},
|
| 3 |
"auto_mapping": null,
|
| 4 |
-
"base_model_name_or_path": "unsloth/
|
| 5 |
"bias": "none",
|
| 6 |
"fan_in_fan_out": false,
|
| 7 |
"inference_mode": true,
|
|
@@ -20,13 +20,13 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
-
"v_proj",
|
| 24 |
-
"down_proj",
|
| 25 |
-
"up_proj",
|
| 26 |
"gate_proj",
|
|
|
|
| 27 |
"k_proj",
|
| 28 |
"q_proj",
|
| 29 |
-
"o_proj"
|
|
|
|
|
|
|
| 30 |
],
|
| 31 |
"task_type": "CAUSAL_LM",
|
| 32 |
"use_dora": false,
|
|
|
|
| 1 |
{
|
| 2 |
"alpha_pattern": {},
|
| 3 |
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "unsloth/llama-3.2-3b-instruct-bnb-4bit",
|
| 5 |
"bias": "none",
|
| 6 |
"fan_in_fan_out": false,
|
| 7 |
"inference_mode": true,
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
|
|
|
|
|
|
|
|
|
| 23 |
"gate_proj",
|
| 24 |
+
"up_proj",
|
| 25 |
"k_proj",
|
| 26 |
"q_proj",
|
| 27 |
+
"o_proj",
|
| 28 |
+
"down_proj",
|
| 29 |
+
"v_proj"
|
| 30 |
],
|
| 31 |
"task_type": "CAUSAL_LM",
|
| 32 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1556140392
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:31d343f6b6b08f7edae93eab588cda7cda9d79bbbaf9fa07d5e1d6c5cb5d3b84
|
| 3 |
size 1556140392
|
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second":
|
| 7 |
-
"total_flos":
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second":
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.178743961352657,
|
| 3 |
+
"eval_loss": 0.015713630244135857,
|
| 4 |
+
"eval_runtime": 20.5637,
|
| 5 |
+
"eval_samples_per_second": 4.863,
|
| 6 |
+
"eval_steps_per_second": 0.146,
|
| 7 |
+
"total_flos": 7.164017972028703e+18,
|
| 8 |
+
"train_loss": 0.021690809104444556,
|
| 9 |
+
"train_runtime": 50941.7483,
|
| 10 |
+
"train_samples_per_second": 4.858,
|
| 11 |
+
"train_steps_per_second": 0.102
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second":
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.178743961352657,
|
| 3 |
+
"eval_loss": 0.015713630244135857,
|
| 4 |
+
"eval_runtime": 20.5637,
|
| 5 |
+
"eval_samples_per_second": 4.863,
|
| 6 |
+
"eval_steps_per_second": 0.146
|
| 7 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.178743961352657,
|
| 3 |
+
"total_flos": 7.164017972028703e+18,
|
| 4 |
+
"train_loss": 0.021690809104444556,
|
| 5 |
+
"train_runtime": 50941.7483,
|
| 6 |
+
"train_samples_per_second": 4.858,
|
| 7 |
+
"train_steps_per_second": 0.102
|
| 8 |
}
|
trainer_log.jsonl
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
trainer_state.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:
|
| 3 |
size 5432
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:feb201ca3a258d1e7b54c4a344a5922ba56be5d3f6ad2abb6d618d250dacc72a
|
| 3 |
size 5432
|
training_eval_loss.png
CHANGED
|
|
training_loss.png
CHANGED
|
|