|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: Cerebras-GPT-1.3B-lora-s-t3000-v300-v1 |
|
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. --> |
|
|
|
# Cerebras-GPT-1.3B-lora-s-t3000-v300-v1 |
|
|
|
This model is a fine-tuned version of [cerebras/Cerebras-GPT-1.3B](https://huggingface.co/cerebras/Cerebras-GPT-1.3B) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.2409 |
|
|
|
## 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: 0.0002 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 16 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 10 |
|
- training_steps: 1000 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 2.4608 | 0.11 | 20 | 2.4030 | |
|
| 2.2475 | 0.21 | 40 | 2.2757 | |
|
| 2.2432 | 0.32 | 60 | 2.2579 | |
|
| 2.3011 | 0.43 | 80 | 2.2467 | |
|
| 2.2293 | 0.53 | 100 | 2.2478 | |
|
| 2.1398 | 0.64 | 120 | 2.2436 | |
|
| 2.2571 | 0.75 | 140 | 2.2413 | |
|
| 2.1577 | 0.85 | 160 | 2.2349 | |
|
| 2.2442 | 0.96 | 180 | 2.2371 | |
|
| 2.2592 | 1.07 | 200 | 2.2342 | |
|
| 2.2082 | 1.17 | 220 | 2.2352 | |
|
| 2.1402 | 1.28 | 240 | 2.2345 | |
|
| 2.1216 | 1.39 | 260 | 2.2345 | |
|
| 2.1758 | 1.49 | 280 | 2.2320 | |
|
| 2.1625 | 1.6 | 300 | 2.2329 | |
|
| 2.1491 | 1.71 | 320 | 2.2311 | |
|
| 2.2307 | 1.81 | 340 | 2.2286 | |
|
| 2.1102 | 1.92 | 360 | 2.2300 | |
|
| 2.2054 | 2.03 | 380 | 2.2278 | |
|
| 2.157 | 2.13 | 400 | 2.2345 | |
|
| 2.0643 | 2.24 | 420 | 2.2359 | |
|
| 2.2134 | 2.35 | 440 | 2.2343 | |
|
| 2.1296 | 2.45 | 460 | 2.2347 | |
|
| 2.1001 | 2.56 | 480 | 2.2346 | |
|
| 2.1401 | 2.67 | 500 | 2.2327 | |
|
| 2.091 | 2.77 | 520 | 2.2328 | |
|
| 2.1365 | 2.88 | 540 | 2.2359 | |
|
| 2.1201 | 2.99 | 560 | 2.2295 | |
|
| 2.1359 | 3.09 | 580 | 2.2338 | |
|
| 2.0979 | 3.2 | 600 | 2.2427 | |
|
| 2.2025 | 3.31 | 620 | 2.2345 | |
|
| 2.1001 | 3.41 | 640 | 2.2368 | |
|
| 2.0228 | 3.52 | 660 | 2.2350 | |
|
| 2.1174 | 3.63 | 680 | 2.2362 | |
|
| 2.0688 | 3.73 | 700 | 2.2372 | |
|
| 2.0368 | 3.84 | 720 | 2.2328 | |
|
| 2.1409 | 3.95 | 740 | 2.2341 | |
|
| 2.0675 | 4.05 | 760 | 2.2377 | |
|
| 2.1805 | 4.16 | 780 | 2.2392 | |
|
| 2.0844 | 4.27 | 800 | 2.2417 | |
|
| 2.0834 | 4.37 | 820 | 2.2395 | |
|
| 2.1396 | 4.48 | 840 | 2.2400 | |
|
| 2.1121 | 4.59 | 860 | 2.2394 | |
|
| 2.0195 | 4.69 | 880 | 2.2391 | |
|
| 2.0564 | 4.8 | 900 | 2.2391 | |
|
| 1.9447 | 4.91 | 920 | 2.2396 | |
|
| 2.2122 | 5.01 | 940 | 2.2384 | |
|
| 2.0482 | 5.12 | 960 | 2.2404 | |
|
| 2.051 | 5.23 | 980 | 2.2411 | |
|
| 2.0345 | 5.33 | 1000 | 2.2409 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.28.0.dev0 |
|
- Pytorch 1.13.1+cu116 |
|
- Datasets 2.11.0 |
|
- Tokenizers 0.13.2 |
|
|
