See axolotl config

axolotl version: 0.4.1

adapter: lora
auto_resume_from_checkpoints: true
base_model: Qwen/Qwen2-0.5B
bf16: auto
chat_template: llama3
dataset_prepared_path: null
dataset_processes: 6
datasets:
- data_files:
  - 67d196c6d8c46ccb_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/67d196c6d8c46ccb_train_data.json
  type:
    field_input: rejected
    field_instruction: prompt
    field_output: chosen
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 3
eval_max_new_tokens: 128
eval_steps: 200
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: error577/6b66dee7-9a0f-41b3-a80f-02c305a544d7
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: 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: 6
mlflow_experiment_name: /tmp/67d196c6d8c46ccb_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
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: 200
sequence_len: 256
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: 21997cd8-7875-4d2d-8537-d9529b0c7b3a
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 21997cd8-7875-4d2d-8537-d9529b0c7b3a
warmup_steps: 30
weight_decay: 0.0
xformers_attention: null

6b66dee7-9a0f-41b3-a80f-02c305a544d7

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: 1.4398

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: 6
• eval_batch_size: 6
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 24
• 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: 30
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
2.0097	0.0004	1	2.0934
1.8783	0.0789	200	1.7714
1.942	0.1577	400	1.6933
1.7195	0.2366	600	1.6511
1.6874	0.3155	800	1.6252
1.637	0.3943	1000	1.5978
1.6431	0.4732	1200	1.5773
1.6009	0.5521	1400	1.5634
1.6767	0.6309	1600	1.5470
1.7032	0.7098	1800	1.5336
1.425	0.7886	2000	1.5209
1.3983	0.8675	2200	1.5104
1.658	0.9464	2400	1.5001
1.2526	1.0252	2600	1.4981
1.3597	1.1041	2800	1.4943
1.502	1.1830	3000	1.4859
1.4823	1.2618	3200	1.4788
1.5438	1.3407	3400	1.4732
1.3918	1.4196	3600	1.4684
1.4004	1.4984	3800	1.4626
1.4335	1.5773	4000	1.4549
1.5331	1.6562	4200	1.4493
1.4744	1.7350	4400	1.4431
1.4191	1.8139	4600	1.4364
1.4737	1.8927	4800	1.4324
1.3473	1.9716	5000	1.4281
1.1141	2.0505	5200	1.4418
1.3193	2.1293	5400	1.4423
1.3628	2.2082	5600	1.4398

Framework versions

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