See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: Qwen/Qwen2.5-1.5B-Instruct
bf16: auto
dataset_prepared_path: null
datasets:
- data_files:
  - f14707e620deedc0_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/f14707e620deedc0_train_data.json
  type:
    field_input: problem
    field_instruction: prompt
    field_output: solution
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 2
flash_attention: null
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: false
group_by_length: false
hub_model_id: sn56a4/dc187fd5-fe0d-409d-9310-bf74d4d1830f
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 16
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
micro_batch_size: 2
mlflow_experiment_name: /tmp/f14707e620deedc0_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: ./outputs/lora-out/taopanda-1_86943aac-324a-4dc3-afc7-638d0cb05ff9
pad_to_sequence_len: null
resume_from_checkpoint: null
sample_packing: false
saves_per_epoch: 1
seed: 3600546949
sequence_len: 2048
shuffle: true
special_tokens: null
strict: false
tf32: false
tokenizer_type: AutoTokenizer
torch_compile: true
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: sn56-miner
wandb_log_model: null
wandb_mode: disabled
wandb_name: null
wandb_project: god
wandb_run: 3l5x
wandb_runid: null
wandb_watch: null
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

dc187fd5-fe0d-409d-9310-bf74d4d1830f

This model is a fine-tuned version of Qwen/Qwen2.5-1.5B-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3676

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: 2
• eval_batch_size: 2
• seed: 3600546949
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• total_eval_batch_size: 4
• 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: 10
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.7195	0.0001	1	0.9369
0.4704	0.5	8603	0.4708
0.2589	1.0	17206	0.3676

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1