File size: 3,889 Bytes
be4bbad |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 |
---
base_model: unsloth/mistral-7b-v0.3-bnb-4bit
library_name: peft
license: apache-2.0
tags:
- unsloth
- generated_from_trainer
model-index:
- name: Mistral-7B-v0.3_metamath_ortho
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# Mistral-7B-v0.3_metamath_ortho
This model is a fine-tuned version of [unsloth/mistral-7b-v0.3-bnb-4bit](https://huggingface.co/unsloth/mistral-7b-v0.3-bnb-4bit) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.6676
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.02
- num_epochs: 1
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.7827 | 0.0211 | 13 | 1.0186 |
| 7.6813 | 0.0421 | 26 | 7.3723 |
| 7.0799 | 0.0632 | 39 | 6.6456 |
| 6.356 | 0.0842 | 52 | 6.2633 |
| 6.2517 | 0.1053 | 65 | 6.2857 |
| 6.2899 | 0.1264 | 78 | 6.2679 |
| 6.285 | 0.1474 | 91 | 6.2945 |
| 6.3073 | 0.1685 | 104 | 6.3867 |
| 6.2797 | 0.1896 | 117 | 6.2024 |
| 6.083 | 0.2106 | 130 | 5.9188 |
| 5.8629 | 0.2317 | 143 | 5.7044 |
| 5.6092 | 0.2527 | 156 | 5.3934 |
| 5.3102 | 0.2738 | 169 | 5.2099 |
| 5.2155 | 0.2949 | 182 | 5.1111 |
| 5.0531 | 0.3159 | 195 | 4.9263 |
| 4.8718 | 0.3370 | 208 | 4.8186 |
| 4.7175 | 0.3580 | 221 | 4.6831 |
| 4.641 | 0.3791 | 234 | 4.6348 |
| 4.5275 | 0.4002 | 247 | 4.5482 |
| 4.4863 | 0.4212 | 260 | 4.4328 |
| 4.4633 | 0.4423 | 273 | 4.3950 |
| 4.4026 | 0.4633 | 286 | 4.3332 |
| 4.3761 | 0.4844 | 299 | 4.2790 |
| 4.2027 | 0.5055 | 312 | 4.1886 |
| 4.1631 | 0.5265 | 325 | 4.1493 |
| 4.0923 | 0.5476 | 338 | 4.1405 |
| 4.1048 | 0.5687 | 351 | 4.0457 |
| 4.0592 | 0.5897 | 364 | 3.9616 |
| 4.0107 | 0.6108 | 377 | 3.9935 |
| 4.021 | 0.6318 | 390 | 3.8987 |
| 3.8899 | 0.6529 | 403 | 3.9228 |
| 3.8158 | 0.6740 | 416 | 3.8781 |
| 3.9124 | 0.6950 | 429 | 3.8955 |
| 3.8687 | 0.7161 | 442 | 3.8612 |
| 3.824 | 0.7371 | 455 | 3.8042 |
| 3.7742 | 0.7582 | 468 | 3.7946 |
| 3.7309 | 0.7793 | 481 | 3.7436 |
| 3.7528 | 0.8003 | 494 | 3.7428 |
| 3.7297 | 0.8214 | 507 | 3.7325 |
| 3.6943 | 0.8424 | 520 | 3.7126 |
| 3.6788 | 0.8635 | 533 | 3.7202 |
| 3.6632 | 0.8846 | 546 | 3.6981 |
| 3.7316 | 0.9056 | 559 | 3.6925 |
| 3.6737 | 0.9267 | 572 | 3.6602 |
| 3.6142 | 0.9478 | 585 | 3.6731 |
| 3.6347 | 0.9688 | 598 | 3.6691 |
| 3.6248 | 0.9899 | 611 | 3.6676 |
### Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.4.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1 |