See axolotl config

axolotl version: 0.4.1

adapter: qlora
base_model: unsloth/SmolLM2-360M-Instruct
bf16: auto
chat_template: llama3
dataloader_num_workers: 6
dataset_prepared_path: null
datasets:
- data_files:
  - 8591808dea314692_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/8591808dea314692_train_data.json
  type:
    field_instruction: instruction
    field_output: output
    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: 50
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 16
gradient_checkpointing: true
group_by_length: true
hub_model_id: error577/0ab8572a-074f-4dff-be51-b76c47b753a5
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0003
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 16
lora_dropout: 0.3
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 8
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 300
micro_batch_size: 1
mlflow_experiment_name: /tmp/8591808dea314692_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 50
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 50
sequence_len: 512
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.02
wandb_entity: null
wandb_mode: online
wandb_name: af92a2e6-f974-41cd-8cc0-cada1f70f806
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: af92a2e6-f974-41cd-8cc0-cada1f70f806
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

0ab8572a-074f-4dff-be51-b76c47b753a5

This model is a fine-tuned version of unsloth/SmolLM2-360M-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.5958

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: 0.0003
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• 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
• training_steps: 300

Training results

Training Loss	Epoch	Step	Validation Loss
1.6213	0.0084	1	2.5594
1.9134	0.4206	50	1.9725
1.7161	0.8412	100	1.7745
1.6513	1.2618	150	1.6839
1.4331	1.6824	200	1.6259
1.5944	2.1030	250	1.5992
1.4112	2.5237	300	1.5958

Framework versions

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