base_model: TinyLlama/TinyLlama_v1.1
datasets:
  - path: mhenrichsen/alpaca_2k_test
    type: alpaca
output_dir: ./outputs/mymodel

sequence_len: 8192
adapter: lora

lora_r: 8
lora_alpha: 16
lora_dropout: 0.05
lora_target_modules:
  - q_proj
  - v_proj
  - k_proj
  - o_proj
  - gate_proj
  - down_proj
  - up_proj

gradient_accumulation_steps: 1
micro_batch_size: 4
num_epochs: 3
optimizer: adamw_bnb_8bit
learning_rate: 2e-5
load_in_8bit: true
train_on_inputs: false
bf16: auto

outputs/mymodel

This model is a fine-tuned version of TinyLlama/TinyLlama_v1.1 on the mhenrichsen/alpaca_2k_test 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: 2e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Use OptimizerNames.ADAMW_BNB 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: 45
num_epochs: 3.0

Training results

Framework versions

PEFT 0.14.0
Transformers 4.49.0
Pytorch 2.5.1+cu124
Datasets 3.2.0
Tokenizers 0.21.0

emmixam/mistral-7b-v0-3-finetune-monbapteme

outputs/mymodel

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

emmixam/
mistral-7b-v0-3-finetune-monbapteme