End of training
Browse files
README.md
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: llama2
|
| 3 |
+
base_model: TheBloke/vigogne-2-70B-chat-GPTQ
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: Vigogne70b-fans
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# Vigogne70b-fans
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [TheBloke/vigogne-2-70B-chat-GPTQ](https://huggingface.co/TheBloke/vigogne-2-70B-chat-GPTQ) on an unknown dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.9593
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 0.0004
|
| 38 |
+
- train_batch_size: 1
|
| 39 |
+
- eval_batch_size: 1
|
| 40 |
+
- seed: 42
|
| 41 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
+
- lr_scheduler_type: linear
|
| 43 |
+
- lr_scheduler_warmup_steps: 2
|
| 44 |
+
- training_steps: 200
|
| 45 |
+
- mixed_precision_training: Native AMP
|
| 46 |
+
|
| 47 |
+
### Training results
|
| 48 |
+
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 51 |
+
| 1.9503 | 0.02 | 10 | 1.6354 |
|
| 52 |
+
| 1.2959 | 0.04 | 20 | 1.2117 |
|
| 53 |
+
| 1.2316 | 0.07 | 30 | 1.1256 |
|
| 54 |
+
| 1.1742 | 0.09 | 40 | 1.0960 |
|
| 55 |
+
| 1.1643 | 0.11 | 50 | 1.0677 |
|
| 56 |
+
| 1.0667 | 0.13 | 60 | 1.0449 |
|
| 57 |
+
| 1.0232 | 0.15 | 70 | 1.0391 |
|
| 58 |
+
| 0.9864 | 0.17 | 80 | 1.0272 |
|
| 59 |
+
| 1.0588 | 0.2 | 90 | 1.0206 |
|
| 60 |
+
| 0.906 | 0.22 | 100 | 1.0020 |
|
| 61 |
+
| 1.098 | 0.24 | 110 | 0.9979 |
|
| 62 |
+
| 0.9973 | 0.26 | 120 | 0.9883 |
|
| 63 |
+
| 0.9999 | 0.28 | 130 | 1.0002 |
|
| 64 |
+
| 1.121 | 0.31 | 140 | 0.9752 |
|
| 65 |
+
| 0.9726 | 0.33 | 150 | 0.9722 |
|
| 66 |
+
| 1.015 | 0.35 | 160 | 0.9680 |
|
| 67 |
+
| 0.8247 | 0.37 | 170 | 0.9664 |
|
| 68 |
+
| 0.823 | 0.39 | 180 | 0.9613 |
|
| 69 |
+
| 0.8921 | 0.41 | 190 | 0.9607 |
|
| 70 |
+
| 1.0024 | 0.44 | 200 | 0.9593 |
|
| 71 |
+
|
| 72 |
+
|
| 73 |
+
### Framework versions
|
| 74 |
+
|
| 75 |
+
- Transformers 4.35.2
|
| 76 |
+
- Pytorch 2.1.0+cu118
|
| 77 |
+
- Datasets 2.15.0
|
| 78 |
+
- Tokenizers 0.15.0
|