File size: 2,666 Bytes
30c78e9 e0f1dcb 30c78e9 e0f1dcb |
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 |
---
library_name: peft
license: apache-2.0
base_model: EleutherAI/pythia-2.8b
tags:
- base_model:adapter:EleutherAI/pythia-2.8b
- lora
- transformers
pipeline_tag: text-generation
model-index:
- name: pythia-2.8b-sft
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. -->
# pythia-2.8b-sft
This model is a fine-tuned version of [EleutherAI/pythia-2.8b](https://huggingface.co/EleutherAI/pythia-2.8b) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.6671
## 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: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 1.0
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 1.8621 | 0.0442 | 100 | 1.7438 |
| 1.7909 | 0.0884 | 200 | 1.7135 |
| 1.7775 | 0.1327 | 300 | 1.7020 |
| 1.7587 | 0.1769 | 400 | 1.6937 |
| 1.7683 | 0.2211 | 500 | 1.6876 |
| 1.7488 | 0.2653 | 600 | 1.6824 |
| 1.7646 | 0.3096 | 700 | 1.6799 |
| 1.7557 | 0.3538 | 800 | 1.6776 |
| 1.7485 | 0.3980 | 900 | 1.6743 |
| 1.7368 | 0.4422 | 1000 | 1.6729 |
| 1.7298 | 0.4865 | 1100 | 1.6705 |
| 1.7525 | 0.5307 | 1200 | 1.6724 |
| 1.7386 | 0.5749 | 1300 | 1.6703 |
| 1.7325 | 0.6191 | 1400 | 1.6684 |
| 1.7306 | 0.6633 | 1500 | 1.6682 |
| 1.7262 | 0.7076 | 1600 | 1.6669 |
| 1.7333 | 0.7518 | 1700 | 1.6675 |
| 1.7318 | 0.7960 | 1800 | 1.6673 |
| 1.7293 | 0.8402 | 1900 | 1.6668 |
| 1.7326 | 0.8845 | 2000 | 1.6671 |
| 1.7378 | 0.9287 | 2100 | 1.6668 |
| 1.7259 | 0.9729 | 2200 | 1.6671 |
### Framework versions
- PEFT 0.17.0
- Transformers 4.55.0
- Pytorch 2.7.1+cu126
- Datasets 4.0.0
- Tokenizers 0.21.4 |