|
--- |
|
base_model: cosmin/falcon-7b-sharded-bf16 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- f1 |
|
model-index: |
|
- name: falcon-7b-sharded |
|
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. --> |
|
|
|
# falcon-7b-sharded |
|
|
|
This model is a fine-tuned version of [cosmin/falcon-7b-sharded-bf16](https://huggingface.co/cosmin/falcon-7b-sharded-bf16) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.3060 |
|
- F1: 0.0027 |
|
|
|
## 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: 0.0001 |
|
- 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_ratio: 0.1 |
|
- training_steps: 500 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:| |
|
| 3.0776 | 1.0 | 55 | 2.8379 | 0.0027 | |
|
| 2.3055 | 1.99 | 110 | 2.5165 | 0.0220 | |
|
| 2.3104 | 2.99 | 165 | 2.4452 | 0.0027 | |
|
| 2.1221 | 4.0 | 221 | 2.3845 | 0.0027 | |
|
| 2.2114 | 5.0 | 276 | 2.3660 | 0.0151 | |
|
| 2.0432 | 5.99 | 331 | 2.3325 | 0.0124 | |
|
| 2.0811 | 6.99 | 386 | 2.3185 | 0.0027 | |
|
| 2.0372 | 8.0 | 442 | 2.3066 | 0.0027 | |
|
| 2.019 | 9.0 | 497 | 2.3058 | 0.0027 | |
|
| 2.0906 | 9.05 | 500 | 2.3060 | 0.0027 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.0 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.14.1 |
|
|
