See axolotl config

axolotl version: 0.4.1

adapter: qlora
auto_resume_from_checkpoints: false
base_model: Qwen/Qwen1.5-0.5B
bf16: auto
chat_template: llama3
dataset_prepared_path: null
dataset_processes: 6
datasets:
- data_files:
  - 0a95f702ed0ba111_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/0a95f702ed0ba111_train_data.json
  type:
    field_instruction: hotel_name
    field_output: review
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 3
eval_max_new_tokens: 128
eval_steps: 500
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
group_by_length: false
hub_model_id: error577/fb0f4949-4c0a-4f03-9c4f-b315d4822a71
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: null
micro_batch_size: 8
mlflow_experiment_name: /tmp/0a95f702ed0ba111_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_torch_4bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 500
sequence_len: 512
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.005
wandb_entity: null
wandb_mode: online
wandb_name: c2e95a13-9e49-4df8-a617-dd6eaea1d861
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: c2e95a13-9e49-4df8-a617-dd6eaea1d861
warmup_steps: 30
weight_decay: 0.0
xformers_attention: null

fb0f4949-4c0a-4f03-9c4f-b315d4822a71

This model is a fine-tuned version of Qwen/Qwen1.5-0.5B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.4562

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
• optimizer: Use OptimizerNames.ADAMW_TORCH_4BIT 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: 30
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
6.2365	0.0000	1	4.9622
3.5179	0.0078	500	3.5677
3.8535	0.0156	1000	3.5265
3.2776	0.0234	1500	3.5129
4.1162	0.0312	2000	3.4760
3.7787	0.0390	2500	3.4721
3.5315	0.0468	3000	3.4617
3.725	0.0546	3500	3.4616
3.3087	0.0624	4000	3.4491
3.3895	0.0702	4500	3.4491
3.4471	0.0780	5000	3.4555
3.5046	0.0858	5500	3.4562

Framework versions

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