---
license: apache-2.0
tags:
- alignment-handbook
- generated_from_trainer
datasets:
- HuggingFaceH4/ultrachat_200k
base_model: EleutherAI/pythia-1.4b
model-index:
- name: pythia-1.4b-sft-full
  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-1.4b-sft-full

This model is a fine-tuned version of [EleutherAI/pythia-1.4b](https://huggingface.co/EleutherAI/pythia-1.4b) on the HuggingFaceH4/ultrachat_200k dataset.
It achieves the following results on the evaluation set:
- Loss: 1.3818

## 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: 12
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 96
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 1.376         | 1.0   | 1335 | 1.3818          |


### Framework versions

- Transformers 4.36.2
- Pytorch 2.2.1
- Datasets 2.14.6
- Tokenizers 0.15.2

# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_nnheui__pythia-1.4b-sft-full)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |34.68|
|AI2 Reasoning Challenge (25-Shot)|32.68|
|HellaSwag (10-Shot)              |52.08|
|MMLU (5-Shot)                    |25.44|
|TruthfulQA (0-shot)              |38.42|
|Winogrande (5-shot)              |57.46|
|GSM8k (5-shot)                   | 1.97|