See axolotl config

axolotl version: 0.4.1

adapter: lora
auto_find_batch_size: true
base_model: unsloth/Qwen2.5-Coder-1.5B-Instruct
bf16: auto
chat_template: llama3
dataloader_num_workers: 12
dataset_prepared_path: null
datasets:
- data_files:
  - 958d1701b8a27f62_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/958d1701b8a27f62_train_data.json
  type:
    field_input: Tags
    field_instruction: Title
    field_output: Predicted_Tags
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 3
early_stopping_threshold: 0.0001
eval_max_new_tokens: 128
eval_steps: 141
eval_strategy: null
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: false
group_by_length: false
hub_model_id: mrferr3t/b4ab149c-1bc1-4a88-bb81-d8ea0279e8b8
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0004
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 141
lora_alpha: 16
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 8
lora_target_linear: true
lr_scheduler: cosine
max_steps: 
micro_batch_size: 32
mlflow_experiment_name: /tmp/958d1701b8a27f62_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 100
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: 
s2_attention: null
sample_packing: false
save_steps: 141
saves_per_epoch: 0
sequence_len: 512
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: 
wandb_name: e3ad808f-bad5-42bb-a274-f81c32f38e91
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: e3ad808f-bad5-42bb-a274-f81c32f38e91
warmup_steps: 100
weight_decay: 0.0
xformers_attention: null

b4ab149c-1bc1-4a88-bb81-d8ea0279e8b8

This model is a fine-tuned version of unsloth/Qwen2.5-Coder-1.5B-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4627

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: 0.0004
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• 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: 100
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0003	1	3.8362
1.5376	0.0428	141	0.7935
0.7066	0.0856	282	0.6471
0.6019	0.1284	423	0.5800
0.5643	0.1712	564	0.5551
0.5421	0.2141	705	0.5317
0.5172	0.2569	846	0.5192
0.515	0.2997	987	0.5122
0.5095	0.3425	1128	0.4982
0.4985	0.3853	1269	0.4995
0.4906	0.4281	1410	0.4858
0.4871	0.4709	1551	0.4803
0.4781	0.5137	1692	0.4793
0.4819	0.5566	1833	0.4732
0.4755	0.5994	1974	0.4731
0.4765	0.6422	2115	0.4681
0.4682	0.6850	2256	0.4665
0.4688	0.7278	2397	0.4600
0.4628	0.7706	2538	0.4619
0.4569	0.8134	2679	0.4605
0.4586	0.8562	2820	0.4627

Framework versions

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