---
license: llama3
base_model: Magpie-Align/Llama-3-8B-Ultrachat-200K
tags:
- axolotl
- generated_from_trainer
model-index:
- name: Llama-3-8B-Ultrachat-200K
  results: []
pipeline_tag: text-generation
---

# QuantFactory/Llama-3-8B-Ultrachat-200K-GGUF
This is quantized version of [Magpie-Align/Llama-3-8B-Ultrachat-200K](https://huggingface.co/Magpie-Align/Llama-3-8B-Ultrachat-200K) created using llama.cpp

# Model Description

[<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)
<details><summary>See axolotl config</summary>

axolotl version: `0.4.0`
```yaml
base_model: meta-llama/Meta-Llama-3-8B
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: PhilipMay/UltraChat-200k-ShareGPT-clean
    type: sharegpt
    conversation: llama3
dataset_prepared_path: last_run_prepared
val_set_size: 0.001
output_dir: ./out_Llama-8B-Ultrachat-200K

sequence_len: 8192
sample_packing: true
eval_sample_packing: false
pad_to_sequence_len: true

wandb_project: SynDa
wandb_entity:
wandb_watch:
wandb_name: Llama-3-8B-Ultrachat-200K
wandb_log_model:
hub_model_id: SynDa/Llama-3-8B-Ultrachat-200K

gradient_accumulation_steps: 8
micro_batch_size: 1
num_epochs: 2
optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 2e-5

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_steps: 100
evals_per_epoch: 1
eval_table_size:
saves_per_epoch: 3
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
  pad_token: <|end_of_text|>

```

</details><br>

# Llama-3-8B-Ultrachat-200K

This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8219

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- total_eval_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 2

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 1.0564        | 0.0010 | 1    | 1.0959          |
| 0.857         | 0.9995 | 1016 | 0.8206          |
| 0.7924        | 1.9785 | 2032 | 0.8219          |


### Framework versions

- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1