See axolotl config

axolotl version: 0.4.1

adapter: lora
auto_resume_from_checkpoints: true
base_model: Qwen/Qwen2.5-0.5B
bf16: auto
chat_template: llama3
dataset_prepared_path: null
dataset_processes: 1
datasets:
- data_files:
  - 6212221035c10067_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/6212221035c10067_train_data.json
  type:
    field_instruction: problem
    field_output: qwq
    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: 200
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: true
group_by_length: false
hub_model_id: error577/f5bdc3f0-9648-414a-b122-d393ffee7a15
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/6212221035c10067_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: 5684560b-1c7a-4061-99f1-c4168d7aa948
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 5684560b-1c7a-4061-99f1-c4168d7aa948
warmup_steps: 30
weight_decay: 0.0
xformers_attention: null

f5bdc3f0-9648-414a-b122-d393ffee7a15

This model is a fine-tuned version of Qwen/Qwen2.5-0.5B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5180

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: 2
• total_train_batch_size: 12
• 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
1.1158	0.0001	1	1.0028
0.7556	0.0196	200	0.6189
0.6999	0.0392	400	0.6003
0.5	0.0588	600	0.5915
0.5623	0.0784	800	0.5847
0.766	0.0979	1000	0.5797
0.5875	0.1175	1200	0.5772
0.6477	0.1371	1400	0.5730
0.5502	0.1567	1600	0.5720
0.7335	0.1763	1800	0.5704
0.6274	0.1959	2000	0.5653
0.5784	0.2155	2200	0.5633
0.6735	0.2351	2400	0.5616
0.6106	0.2547	2600	0.5594
0.5109	0.2742	2800	0.5588
0.6675	0.2938	3000	0.5574
0.5951	0.3134	3200	0.5555
0.6282	0.3330	3400	0.5538
0.8245	0.3526	3600	0.5513
0.5453	0.3722	3800	0.5518
0.6636	0.3918	4000	0.5494
0.5605	0.4114	4200	0.5490
0.6634	0.4310	4400	0.5470
0.5609	0.4505	4600	0.5461
0.8304	0.4701	4800	0.5442
0.609	0.4897	5000	0.5438
0.4672	0.5093	5200	0.5421
0.6153	0.5289	5400	0.5431
0.4617	0.5485	5600	0.5427
0.6679	0.5681	5800	0.5400
0.6296	0.5877	6000	0.5386
0.773	0.6072	6200	0.5386
0.4485	0.6268	6400	0.5366
0.5028	0.6464	6600	0.5354
0.5002	0.6660	6800	0.5326
0.6002	0.6856	7000	0.5329
0.4106	0.7052	7200	0.5340
0.5775	0.7248	7400	0.5319
0.5416	0.7444	7600	0.5311
0.6253	0.7640	7800	0.5307
0.6384	0.7835	8000	0.5293
0.6838	0.8031	8200	0.5274
0.5797	0.8227	8400	0.5248
0.614	0.8423	8600	0.5252
0.6764	0.8619	8800	0.5239
0.662	0.8815	9000	0.5230
0.5421	0.9011	9200	0.5223
0.6374	0.9207	9400	0.5217
0.5016	0.9403	9600	0.5214
0.4231	0.9598	9800	0.5177
0.5269	0.9794	10000	0.5164
0.4963	0.9990	10200	0.5167
0.5594	1.0186	10400	0.5190
0.5187	1.0382	10600	0.5180

Framework versions

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