Commit ·
2f4038e
1
Parent(s): 5aa36ba
End of training
Browse files
README.md
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Llama-2-7b-hf
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: llama2_lora_ice2
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# llama2_lora_ice2
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on an unknown dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 2.1126
|
| 18 |
+
|
| 19 |
+
## Model description
|
| 20 |
+
|
| 21 |
+
More information needed
|
| 22 |
+
|
| 23 |
+
## Intended uses & limitations
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Training and evaluation data
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training procedure
|
| 32 |
+
|
| 33 |
+
### Training hyperparameters
|
| 34 |
+
|
| 35 |
+
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 0.0001
|
| 37 |
+
- train_batch_size: 8
|
| 38 |
+
- eval_batch_size: 8
|
| 39 |
+
- seed: 42
|
| 40 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 41 |
+
- lr_scheduler_type: linear
|
| 42 |
+
- training_steps: 100000
|
| 43 |
+
|
| 44 |
+
### Training results
|
| 45 |
+
|
| 46 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 47 |
+
|:-------------:|:-----:|:------:|:---------------:|
|
| 48 |
+
| 2.3504 | 0.19 | 5000 | 2.3109 |
|
| 49 |
+
| 2.3206 | 0.38 | 10000 | 2.2213 |
|
| 50 |
+
| 2.2321 | 0.57 | 15000 | 2.1817 |
|
| 51 |
+
| 2.1524 | 0.75 | 20000 | 2.1579 |
|
| 52 |
+
| 2.2333 | 0.94 | 25000 | 2.1455 |
|
| 53 |
+
| 2.1409 | 1.13 | 30000 | 2.1367 |
|
| 54 |
+
| 2.1875 | 1.32 | 35000 | 2.1301 |
|
| 55 |
+
| 2.1823 | 1.51 | 40000 | 2.1254 |
|
| 56 |
+
| 2.1062 | 1.7 | 45000 | 2.1217 |
|
| 57 |
+
| 2.1374 | 1.88 | 50000 | 2.1192 |
|
| 58 |
+
| 2.1645 | 2.07 | 55000 | 2.1174 |
|
| 59 |
+
| 2.1595 | 2.26 | 60000 | 2.1159 |
|
| 60 |
+
| 2.1405 | 2.45 | 65000 | 2.1149 |
|
| 61 |
+
| 2.1585 | 2.64 | 70000 | 2.1139 |
|
| 62 |
+
| 2.1423 | 2.83 | 75000 | 2.1132 |
|
| 63 |
+
| 2.1656 | 3.02 | 80000 | 2.1129 |
|
| 64 |
+
| 2.1159 | 3.2 | 85000 | 2.1128 |
|
| 65 |
+
| 2.1565 | 3.39 | 90000 | 2.1127 |
|
| 66 |
+
| 2.115 | 3.58 | 95000 | 2.1126 |
|
| 67 |
+
| 2.144 | 3.77 | 100000 | 2.1126 |
|
| 68 |
+
|
| 69 |
+
|
| 70 |
+
### Framework versions
|
| 71 |
+
|
| 72 |
+
- Transformers 4.34.1
|
| 73 |
+
- Pytorch 2.0.1+cu117
|
| 74 |
+
- Datasets 2.14.5
|
| 75 |
+
- Tokenizers 0.14.1
|