h0m4m/sympter-1

See axolotl config

axolotl version: 0.8.0.dev0

adapter: lora
base_model: deepseek-ai/DeepSeek-R1-Distill-Llama-8B
bf16: auto

datasets:
  - path: train.json
    type: completion

val_set_size: 0
sequence_len: 4096
sample_packing: true

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

output_dir: ./outputs/mymodel

load_in_8bit: false
train_on_inputs: false
save_safetensors: true
save_only_model: true
flash_attention: true
logging_steps: 10

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

outputs/mymodel

This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Llama-8B on the train.json 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: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 8
• 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: 16
• num_epochs: 1.0

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