llama-3.1-8b-email-writer-axolotl

<!-- 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/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>

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

axolotl version: 0.13.0.dev0

base_model: NousResearch/Meta-Llama-3.1-8B
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer

load_in_4bit: true
strict: false

chat_template: llama3
datasets:
  - path: rizoner/email-training-data
    type: alpaca

dataset_prepared_path: null
val_set_size: 0.0
output_dir: ./outputs/email-writer

sequence_len: 1024
sample_packing: false
pad_to_sequence_len: false

adapter: lora
lora_r: 16
lora_alpha: 8
lora_dropout: 0.05
lora_target_modules:
  - q_proj
  - v_proj

num_epochs: 6
gradient_accumulation_steps: 16
micro_batch_size: 1

optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 0.0002

train_on_inputs: false
group_by_length: false
bf16: false
fp16: true
tf32: false

gradient_checkpointing: true

flash_attention: false

warmup_ratio: 0.1
evaluation_strategy: "no"
save_strategy: epoch
save_total_limit: 1

logging_steps: 1

special_tokens:
  pad_token: "<|end_of_text|>"

hub_model_id: rizoner/llama-3.1-8b-email-writer-axolotl
push_to_hub: true
hf_use_auth_token: true

</details><br>

llama-3.1-8b-email-writer-axolotl

This model is a fine-tuned version of NousResearch/Meta-Llama-3.1-8B on the rizoner/email-training-data dataset.

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: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT 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: 39
• training_steps: 394
• mixed_precision_training: Native AMP

Training results

Framework versions

• PEFT 0.17.1
• Transformers 4.57.1
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1