See axolotl config

axolotl version: 0.3.0

base_model: openaccess-ai-collective/tiny-mistral
flash_attention: true
sequence_len: 1024
load_in_8bit: true
adapter: lora
lora_r: 32
lora_alpha: 64
lora_dropout: 0.05
lora_target_linear: true
val_set_size: 0.1
special_tokens:
  unk_token: <unk>
  bos_token: <s>
  eos_token: </s>
datasets:
  - path: mhenrichsen/alpaca_2k_test
    type: alpaca
num_epochs: 2
micro_batch_size: 2
gradient_accumulation_steps: 1
output_dir: temp_dir
learning_rate: 0.00001
optimizer: adamw_torch
lr_scheduler: cosine
max_steps: 20
save_steps: 10
eval_steps: 10
hub_model_id: hamel/axolotl-test
dataset_processes: 1

axolotl-test

This model is a fine-tuned version of openaccess-ai-collective/tiny-mistral on the None 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: 1e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• training_steps: 20

Framework versions

• Transformers 4.37.0.dev0
• Pytorch 2.1.0
• 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: None
• 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