See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: unsloth/gemma-2-9b-it
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - dfb6bcb29a5bcb68_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/dfb6bcb29a5bcb68_train_data.json
  type:
    field_input: statements
    field_instruction: quiz
    field_output: solution_text
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
device_map:
  ? ''
  : 0,1,2,3,4,5,6,7
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 400
eval_table_size: null
flash_attention: false
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: Alphatao/9529a318-6254-4cce-a04e-f3fe0831cffc
hub_repo: null
hub_strategy: null
hub_token: null
learning_rate: 0.0002
load_best_model_at_end: true
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lora_target_modules:
- q_proj
- k_proj
- v_proj
- o_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 4599
micro_batch_size: 2
mlflow_experiment_name: /tmp/dfb6bcb29a5bcb68_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
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: 400
sequence_len: 1024
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.04
wandb_entity: null
wandb_mode: online
wandb_name: a0d4f75b-7403-424e-b96b-ba4e344fa29f
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: a0d4f75b-7403-424e-b96b-ba4e344fa29f
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

9529a318-6254-4cce-a04e-f3fe0831cffc

This model is a fine-tuned version of unsloth/gemma-2-9b-it on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0265

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: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• 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: 4599

Training results

Training Loss	Epoch	Step	Validation Loss
1.2063	0.0002	1	1.1616
0.106	0.0920	400	0.0863
0.0832	0.1841	800	0.0718
0.1227	0.2761	1200	0.0985
0.0634	0.3682	1600	0.0632
0.0363	0.4602	2000	0.0509
0.0411	0.5523	2400	0.0454
0.067	0.6443	2800	0.0390
0.056	0.7364	3200	0.0350
0.0234	0.8284	3600	0.0316
0.0204	0.9204	4000	0.0272
0.013	1.0125	4400	0.0265

Framework versions

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