---
library_name: peft
tags:
- generated_from_trainer
base_model: awilliamson/tinyllama-slider
model-index:
- name: empathy-lora-out
  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. -->

[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
# empathy-lora-out

This model is a fine-tuned version of [awilliamson/tinyllama-slider](https://huggingface.co/awilliamson/tinyllama-slider) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2988

## 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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 25
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 1.5961        | 0.03  | 1    | 1.5286          |
| 1.6946        | 0.27  | 8    | 1.5283          |
| 1.5327        | 0.54  | 16   | 1.5186          |
| 1.4197        | 0.81  | 24   | 1.4726          |
| 1.4979        | 1.03  | 32   | 1.4323          |
| 1.3198        | 1.3   | 40   | 1.3997          |
| 1.4632        | 1.57  | 48   | 1.3762          |
| 1.3486        | 1.84  | 56   | 1.3547          |
| 1.3901        | 2.08  | 64   | 1.3386          |
| 1.3964        | 2.35  | 72   | 1.3291          |
| 1.3844        | 2.62  | 80   | 1.3184          |
| 1.2799        | 2.89  | 88   | 1.3108          |
| 1.3213        | 3.13  | 96   | 1.3061          |
| 1.0652        | 3.39  | 104  | 1.3050          |
| 1.1837        | 3.66  | 112  | 1.3024          |
| 1.4713        | 3.93  | 120  | 1.2992          |
| 1.3074        | 4.17  | 128  | 1.2979          |
| 1.2039        | 4.44  | 136  | 1.2993          |
| 1.1574        | 4.71  | 144  | 1.2988          |


### Framework versions

- Transformers 4.37.0.dev0
- Pytorch 2.0.1
- Datasets 2.15.0
- Tokenizers 0.15.0
## Training procedure


The following `bitsandbytes` quantization config was used during training:
- quant_method: bitsandbytes
- load_in_8bit: True
- load_in_4bit: False
- llm_int8_threshold: 6.0
- llm_int8_skip_modules: None
- llm_int8_enable_fp32_cpu_offload: False
- llm_int8_has_fp16_weight: False
- bnb_4bit_quant_type: fp4
- bnb_4bit_use_double_quant: False
- bnb_4bit_compute_dtype: float32

### Framework versions


- PEFT 0.6.0