Commit
•
a2a4b31
1
Parent(s):
44f59c8
Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Llama-3.1-8B
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: llama3.1
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: Llama-3.1-8B-medquad-V2
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# Llama-3.1-8B-medquad-V2
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.8959
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0002
|
| 41 |
+
- train_batch_size: 16
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
+
- seed: 42
|
| 44 |
+
- gradient_accumulation_steps: 12
|
| 45 |
+
- total_train_batch_size: 192
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: reduce_lr_on_plateau
|
| 48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
+
- num_epochs: 7
|
| 50 |
+
- mixed_precision_training: Native AMP
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 56 |
+
| 1.2503 | 0.1462 | 10 | 1.1359 |
|
| 57 |
+
| 1.1182 | 0.2923 | 20 | 1.0199 |
|
| 58 |
+
| 1.0864 | 0.4385 | 30 | 0.9856 |
|
| 59 |
+
| 0.9031 | 0.5847 | 40 | 0.9681 |
|
| 60 |
+
| 1.0773 | 0.7308 | 50 | 0.9499 |
|
| 61 |
+
| 0.9575 | 0.8770 | 60 | 0.9427 |
|
| 62 |
+
| 0.9768 | 1.0231 | 70 | 0.9452 |
|
| 63 |
+
| 0.9673 | 1.1693 | 80 | 0.9264 |
|
| 64 |
+
| 0.8541 | 1.3155 | 90 | 0.9282 |
|
| 65 |
+
| 0.9772 | 1.4616 | 100 | 0.9180 |
|
| 66 |
+
| 0.8427 | 1.6078 | 110 | 0.9211 |
|
| 67 |
+
| 0.9317 | 1.7540 | 120 | 0.9142 |
|
| 68 |
+
| 0.9498 | 1.9001 | 130 | 0.9011 |
|
| 69 |
+
| 0.8412 | 2.0463 | 140 | 0.9036 |
|
| 70 |
+
| 0.899 | 2.1924 | 150 | 0.9031 |
|
| 71 |
+
| 0.7488 | 2.3386 | 160 | 0.8990 |
|
| 72 |
+
| 0.8824 | 2.4848 | 170 | 0.9033 |
|
| 73 |
+
| 0.8334 | 2.6309 | 180 | 0.8959 |
|
| 74 |
+
|
| 75 |
+
|
| 76 |
+
### Framework versions
|
| 77 |
+
|
| 78 |
+
- PEFT 0.13.0
|
| 79 |
+
- Transformers 4.45.1
|
| 80 |
+
- Pytorch 2.4.1+cu121
|
| 81 |
+
- Datasets 3.0.1
|
| 82 |
+
- Tokenizers 0.20.0
|