yangming0121/initial_trained_with_math_data

See axolotl config

axolotl version: 0.10.0.dev0

base_model: deepseek-ai/deepseek-llm-7b-base

load_in_8bit: false
load_in_4bit: false
strict: false

chat_template: deepseek_v2
datasets:
  - path: microsoft/orca-math-word-problems-200k
    type:
      system_prompt: ""

      field_system:
      field_instruction: question
      field_input:
      field_output: answer
      format: "[INST] {instruction} [/INST]"
      no_input_format: "[INST] {instruction} [/INST]"
      
    split: train[:80%]

val_set_size: 0.0
output_dir: ./outputs
save_safetensors: true

sequence_len: 4096
sample_packing: false
pad_to_sequence_len: true

gradient_accumulation_steps: 2
micro_batch_size: 8
num_epochs: 4
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 0.0002

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

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: true
logging_steps: 1
flash_attention: true

warmup_steps: 500
evals_per_epoch: 1
eval_table_size: 20
eval_max_new_tokens: 128
saves_per_epoch: 1
save_total_limit: 10

outputs

This model is a fine-tuned version of deepseek-ai/deepseek-llm-7b-base on the microsoft/orca-math-word-problems-200k 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: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• total_eval_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH 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: 500
• training_steps: 5001

Training results

Framework versions

• Transformers 4.52.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1