--- library_name: peft license: other base_model: Qwen/Qwen1.5-1.8B tags: - axolotl - generated_from_trainer model-index: - name: 1fb8a15e-fa91-47a7-8821-aa3289a37f7b results: [] --- [Built with Axolotl](https://github.com/axolotl-ai-cloud/axolotl)
See axolotl config axolotl version: `0.4.1` ```yaml adapter: lora base_model: Qwen/Qwen1.5-1.8B bf16: auto chat_template: llama3 cosine_min_lr_ratio: 0.1 data_processes: 4 dataset_prepared_path: null datasets: - data_files: - 092c162f01e06512_train_data.json ds_type: json format: custom num_proc: 4 path: /workspace/input_data/092c162f01e06512_train_data.json streaming: true type: field_instruction: smiles field_output: inchi format: '{instruction}' no_input_format: '{instruction}' system_format: '{system}' system_prompt: '' debug: null deepspeed: null device_map: lm_head: 3 model.embed_tokens: 0 model.layers.0: 0 model.layers.1: 0 model.layers.10: 3 model.layers.11: 3 model.layers.2: 0 model.layers.3: 1 model.layers.4: 1 model.layers.5: 1 model.layers.6: 2 model.layers.7: 2 model.layers.8: 2 model.layers.9: 3 model.norm: 3 do_eval: true early_stopping_patience: 1 eval_batch_size: 1 eval_sample_packing: false eval_steps: 25 evaluation_strategy: steps flash_attention: false fp16: null fsdp: null fsdp_config: null gradient_accumulation_steps: 32 gradient_checkpointing: true group_by_length: true hub_model_id: dsakerkwq/1fb8a15e-fa91-47a7-8821-aa3289a37f7b hub_strategy: checkpoint hub_token: null learning_rate: 0.0001 load_in_4bit: false load_in_8bit: false local_rank: null logging_steps: 1 lora_alpha: 64 lora_dropout: 0.05 lora_fan_in_fan_out: null lora_model_dir: null lora_r: 32 lora_target_linear: true lora_target_modules: - q_proj - v_proj lr_scheduler: cosine max_grad_norm: 0.3 max_memory: 0: 60GB 1: 70GB 2: 70GB 3: 70GB cpu: 96GB max_steps: 50 micro_batch_size: 1 mixed_precision: bf16 mlflow_experiment_name: /tmp/092c162f01e06512_train_data.json model_type: AutoModelForCausalLM num_epochs: 3 optim_args: adam_beta1: 0.9 adam_beta2: 0.95 adam_epsilon: 1e-5 optimizer: adamw_torch output_dir: miner_id_24 pad_to_sequence_len: true resume_from_checkpoint: null s2_attention: null sample_packing: false save_steps: 25 save_strategy: steps sequence_len: 2048 strict: false tf32: false tokenizer_type: AutoTokenizer torch_compile: false torch_dtype: bfloat16 train_on_inputs: false trust_remote_code: true use_cache: false val_set_size: 50 wandb_entity: null wandb_mode: online wandb_name: 1fb8a15e-fa91-47a7-8821-aa3289a37f7b wandb_project: Public_TuningSN wandb_runid: 1fb8a15e-fa91-47a7-8821-aa3289a37f7b warmup_ratio: 0.05 weight_decay: 0.01 xformers_attention: null ```

# 1fb8a15e-fa91-47a7-8821-aa3289a37f7b This model is a fine-tuned version of [Qwen/Qwen1.5-1.8B](https://huggingface.co/Qwen/Qwen1.5-1.8B) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.6595 ## 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: 0.0001 - train_batch_size: 1 - eval_batch_size: 1 - seed: 42 - distributed_type: multi-GPU - num_devices: 4 - gradient_accumulation_steps: 32 - total_train_batch_size: 128 - total_eval_batch_size: 4 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5 - lr_scheduler_type: cosine - lr_scheduler_warmup_steps: 2 - training_steps: 50 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 1.9653 | 0.0010 | 1 | 2.1729 | | 0.7289 | 0.0239 | 25 | 0.7704 | | 0.7522 | 0.0479 | 50 | 0.6595 | ### Framework versions - PEFT 0.13.2 - Transformers 4.46.0 - Pytorch 2.5.0+cu124 - Datasets 3.0.1 - Tokenizers 0.20.1