fine-tuning-Phi2-with-webglm-qa-with-lora_9
Browse files- README.md +132 -0
- adapter_config.json +30 -0
- adapter_model.safetensors +3 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,132 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
base_model: microsoft/phi-2
|
| 7 |
+
model-index:
|
| 8 |
+
- name: fine-tuning-Phi2-with-webglm-qa-with-lora_9
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# fine-tuning-Phi2-with-webglm-qa-with-lora_9
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.1781
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 5e-05
|
| 39 |
+
- train_batch_size: 2
|
| 40 |
+
- eval_batch_size: 2
|
| 41 |
+
- seed: 42
|
| 42 |
+
- gradient_accumulation_steps: 5
|
| 43 |
+
- total_train_batch_size: 10
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: linear
|
| 46 |
+
- lr_scheduler_warmup_steps: 60
|
| 47 |
+
- training_steps: 700
|
| 48 |
+
- mixed_precision_training: Native AMP
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 54 |
+
| 1.9045 | 0.16 | 10 | 1.6827 |
|
| 55 |
+
| 1.6798 | 0.32 | 20 | 1.5996 |
|
| 56 |
+
| 1.5222 | 0.48 | 30 | 1.4548 |
|
| 57 |
+
| 1.286 | 0.64 | 40 | 1.2269 |
|
| 58 |
+
| 1.0827 | 0.8 | 50 | 1.1089 |
|
| 59 |
+
| 0.9913 | 0.96 | 60 | 1.0021 |
|
| 60 |
+
| 0.881 | 1.13 | 70 | 0.8796 |
|
| 61 |
+
| 0.7673 | 1.29 | 80 | 0.7637 |
|
| 62 |
+
| 0.6315 | 1.45 | 90 | 0.6618 |
|
| 63 |
+
| 0.554 | 1.61 | 100 | 0.5964 |
|
| 64 |
+
| 0.5132 | 1.77 | 110 | 0.5487 |
|
| 65 |
+
| 0.4915 | 1.93 | 120 | 0.5030 |
|
| 66 |
+
| 0.4787 | 2.09 | 130 | 0.4705 |
|
| 67 |
+
| 0.4298 | 2.25 | 140 | 0.4451 |
|
| 68 |
+
| 0.4009 | 2.41 | 150 | 0.4099 |
|
| 69 |
+
| 0.3886 | 2.57 | 160 | 0.3889 |
|
| 70 |
+
| 0.3729 | 2.73 | 170 | 0.3674 |
|
| 71 |
+
| 0.3236 | 2.89 | 180 | 0.3527 |
|
| 72 |
+
| 0.3377 | 3.05 | 190 | 0.3407 |
|
| 73 |
+
| 0.3356 | 3.22 | 200 | 0.3261 |
|
| 74 |
+
| 0.3083 | 3.38 | 210 | 0.3121 |
|
| 75 |
+
| 0.2794 | 3.54 | 220 | 0.2992 |
|
| 76 |
+
| 0.2917 | 3.7 | 230 | 0.2926 |
|
| 77 |
+
| 0.2895 | 3.86 | 240 | 0.2879 |
|
| 78 |
+
| 0.2764 | 4.02 | 250 | 0.2782 |
|
| 79 |
+
| 0.2585 | 4.18 | 260 | 0.2732 |
|
| 80 |
+
| 0.2489 | 4.34 | 270 | 0.2678 |
|
| 81 |
+
| 0.2401 | 4.5 | 280 | 0.2591 |
|
| 82 |
+
| 0.2489 | 4.66 | 290 | 0.2573 |
|
| 83 |
+
| 0.2529 | 4.82 | 300 | 0.2501 |
|
| 84 |
+
| 0.2637 | 4.98 | 310 | 0.2455 |
|
| 85 |
+
| 0.255 | 5.14 | 320 | 0.2411 |
|
| 86 |
+
| 0.2266 | 5.31 | 330 | 0.2370 |
|
| 87 |
+
| 0.2209 | 5.47 | 340 | 0.2326 |
|
| 88 |
+
| 0.2311 | 5.63 | 350 | 0.2276 |
|
| 89 |
+
| 0.2203 | 5.79 | 360 | 0.2275 |
|
| 90 |
+
| 0.2048 | 5.95 | 370 | 0.2210 |
|
| 91 |
+
| 0.2133 | 6.11 | 380 | 0.2179 |
|
| 92 |
+
| 0.2045 | 6.27 | 390 | 0.2142 |
|
| 93 |
+
| 0.2053 | 6.43 | 400 | 0.2137 |
|
| 94 |
+
| 0.1898 | 6.59 | 410 | 0.2102 |
|
| 95 |
+
| 0.1897 | 6.75 | 420 | 0.2073 |
|
| 96 |
+
| 0.2141 | 6.91 | 430 | 0.2040 |
|
| 97 |
+
| 0.1872 | 7.07 | 440 | 0.2028 |
|
| 98 |
+
| 0.1938 | 7.23 | 450 | 0.1998 |
|
| 99 |
+
| 0.187 | 7.4 | 460 | 0.2004 |
|
| 100 |
+
| 0.1782 | 7.56 | 470 | 0.1973 |
|
| 101 |
+
| 0.1908 | 7.72 | 480 | 0.1967 |
|
| 102 |
+
| 0.1899 | 7.88 | 490 | 0.1912 |
|
| 103 |
+
| 0.1823 | 8.04 | 500 | 0.1912 |
|
| 104 |
+
| 0.1769 | 8.2 | 510 | 0.1915 |
|
| 105 |
+
| 0.1774 | 8.36 | 520 | 0.1909 |
|
| 106 |
+
| 0.1793 | 8.52 | 530 | 0.1890 |
|
| 107 |
+
| 0.1853 | 8.68 | 540 | 0.1880 |
|
| 108 |
+
| 0.1785 | 8.84 | 550 | 0.1861 |
|
| 109 |
+
| 0.1515 | 9.0 | 560 | 0.1845 |
|
| 110 |
+
| 0.1689 | 9.16 | 570 | 0.1845 |
|
| 111 |
+
| 0.1552 | 9.32 | 580 | 0.1836 |
|
| 112 |
+
| 0.1712 | 9.49 | 590 | 0.1828 |
|
| 113 |
+
| 0.1642 | 9.65 | 600 | 0.1818 |
|
| 114 |
+
| 0.1703 | 9.81 | 610 | 0.1806 |
|
| 115 |
+
| 0.1772 | 9.97 | 620 | 0.1804 |
|
| 116 |
+
| 0.1615 | 10.13 | 630 | 0.1796 |
|
| 117 |
+
| 0.1494 | 10.29 | 640 | 0.1801 |
|
| 118 |
+
| 0.1702 | 10.45 | 650 | 0.1798 |
|
| 119 |
+
| 0.1656 | 10.61 | 660 | 0.1787 |
|
| 120 |
+
| 0.1688 | 10.77 | 670 | 0.1782 |
|
| 121 |
+
| 0.1452 | 10.93 | 680 | 0.1780 |
|
| 122 |
+
| 0.1732 | 11.09 | 690 | 0.1782 |
|
| 123 |
+
| 0.1719 | 11.25 | 700 | 0.1781 |
|
| 124 |
+
|
| 125 |
+
|
| 126 |
+
### Framework versions
|
| 127 |
+
|
| 128 |
+
- PEFT 0.7.1
|
| 129 |
+
- Transformers 4.36.2
|
| 130 |
+
- Pytorch 2.0.0
|
| 131 |
+
- Datasets 2.15.0
|
| 132 |
+
- Tokenizers 0.15.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "microsoft/phi-2",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
+
"lora_alpha": 32,
|
| 13 |
+
"lora_dropout": 0.05,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 16,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"fc2",
|
| 23 |
+
"v_proj",
|
| 24 |
+
"dense",
|
| 25 |
+
"k_proj",
|
| 26 |
+
"fc1",
|
| 27 |
+
"q_proj"
|
| 28 |
+
],
|
| 29 |
+
"task_type": "CAUSAL_LM"
|
| 30 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:97f58fc27095c6325287ffdce28bd8f327a522a28bf57e35c3fe58c4548e4bf1
|
| 3 |
+
size 94422368
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f59b90fbb1bd20e2d5185048b0170c685bd4e3b401e04deff1c3b899d51422bd
|
| 3 |
+
size 4283
|