See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: unsloth/Qwen2.5-Coder-1.5B-Instruct
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 26e357e10ca56cb1_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/26e357e10ca56cb1_train_data.json
  type:
    field_input: input
    field_instruction: instruction
    field_output: output
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 10
eval_table_size: null
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: false
group_by_length: false
hub_model_id: mrferr3t/73404e5a-0e16-4392-897f-478526658a46
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0005
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
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: 100
micro_batch_size: 2
mlflow_experiment_name: /tmp/26e357e10ca56cb1_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: 20
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: online
wandb_name: f2373d9f-8e95-4acb-8856-941411fbcc10
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: f2373d9f-8e95-4acb-8856-941411fbcc10
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

73404e5a-0e16-4392-897f-478526658a46

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: 1.6609

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: 0.0005
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Use adamw_bnb_8bit 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: 10
• training_steps: 100

Training results

Training Loss	Epoch	Step	Validation Loss
2.4987	0.0002	1	2.7995
2.3503	0.0016	10	2.4840
2.0748	0.0033	20	2.2137
1.8103	0.0049	30	2.0079
2.0097	0.0065	40	1.8863
1.868	0.0081	50	1.8088
1.6384	0.0098	60	1.7546
1.7302	0.0114	70	1.7126
1.715	0.0130	80	1.6765
1.6301	0.0146	90	1.6636
1.4097	0.0163	100	1.6609

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.3.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.1