See axolotl config

axolotl version: 0.4.1

adapter: lora
auto_find_batch_size: true
base_model: unsloth/Qwen2.5-14B
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - f64bc0f2d1a0f0fe_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/f64bc0f2d1a0f0fe_train_data.json
  type:
    field_input: rc_name
    field_instruction: QA_type
    field_output: context
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
do_eval: true
early_stopping_patience: 3
eval_max_new_tokens: 128
eval_steps: 50
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: false
group_by_length: true
hub_model_id: lesso13/b81b7130-610a-483a-ac21-8141a5d8cf0c
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.000213
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 10
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 500
micro_batch_size: 4
mlflow_experiment_name: /tmp/f64bc0f2d1a0f0fe_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
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: 50
saves_per_epoch: null
seed: 130
sequence_len: 512
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: e14a5b87-8459-454e-a245-5a47cffc4bfe
wandb_project: 13a
wandb_run: your_name
wandb_runid: e14a5b87-8459-454e-a245-5a47cffc4bfe
warmup_steps: 50
weight_decay: 0.0
xformers_attention: null

b81b7130-610a-483a-ac21-8141a5d8cf0c

This model is a fine-tuned version of unsloth/Qwen2.5-14B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7122

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: 0.000213
• train_batch_size: 4
• eval_batch_size: 4
• seed: 130
• gradient_accumulation_steps: 2
• 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: 50
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0004	1	1.0486
0.4496	0.0217	50	0.7273
0.8438	0.0434	100	0.7393
0.4888	0.0651	150	0.7212
0.9703	0.0868	200	0.7333
0.7938	0.1085	250	0.7188
0.8343	0.1302	300	0.7189
0.5117	0.1519	350	0.7170
0.9712	0.1736	400	0.7105
0.6231	0.1953	450	0.7127
0.8418	0.2170	500	0.7122

Framework versions

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