See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: Qwen/Qwen2-0.5B
bf16: true
chat_template: llama3
dataloader_num_workers: 12
dataset_prepared_path: null
datasets:
- data_files:
  - 5513b998661e6cb3_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/5513b998661e6cb3_train_data.json
  type:
    field_input: old_contents
    field_instruction: instruction
    field_output: content
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map: auto
do_eval: true
early_stopping_patience: 3
eval_batch_size: 4
eval_max_new_tokens: 128
eval_steps: 300
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: false
group_by_length: true
hub_model_id: aleegis12/c4b0231e-c4ce-45ca-acc6-2b558aff3187
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 10
lora_alpha: 128
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 6000
micro_batch_size: 4
mlflow_experiment_name: /tmp/5513b998661e6cb3_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 10
optim_args:
  adam_beta1: 0.9
  adam_beta2: 0.999
  adam_epsilon: 1e-8
optimizer: adamw_torch_fused
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 300
saves_per_epoch: null
sequence_len: 1024
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: cdec897c-7948-4e87-9bae-779bd9138cdd
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: cdec897c-7948-4e87-9bae-779bd9138cdd
warmup_steps: 100
weight_decay: 0.1
xformers_attention: null

c4b0231e-c4ce-45ca-acc6-2b558aff3187

This model is a fine-tuned version of Qwen/Qwen2-0.5B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1922

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.999,adam_epsilon=1e-8
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• training_steps: 6000

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0008	1	0.4837
0.3095	0.2265	300	0.1902
0.3482	0.4531	600	0.1907
0.2556	0.6796	900	0.1902
0.286	0.9062	1200	0.1882
0.1485	1.1327	1500	0.1917
0.1464	1.3593	1800	0.1934
0.1351	1.5858	2100	0.1922

Framework versions

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