See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: unsloth/Qwen2.5-1.5B
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 4ce2c91eec978c10_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/4ce2c91eec978c10_train_data.json
  type:
    field_input: context
    field_instruction: statement
    field_output: justification
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 30
eval_max_new_tokens: 128
eval_steps: 50
eval_table_size: null
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 16
gradient_checkpointing: true
group_by_length: false
hub_model_id: Romain-XV/b6b8192a-ae38-4b6c-8d1d-38e19cbf1cb6
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: 32
lora_dropout: 0.05
lora_fan_in_fan_out: true
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lora_target_modules:
- q_proj
- k_proj
- v_proj
lr_scheduler: cosine
micro_batch_size: 4
mlflow_experiment_name: /tmp/4ce2c91eec978c10_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
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: 100
sequence_len: 1024
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: d3d838cc-f27d-4fe4-bdd1-e56818955b47
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: d3d838cc-f27d-4fe4-bdd1-e56818955b47
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

b6b8192a-ae38-4b6c-8d1d-38e19cbf1cb6

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

• Loss: 2.0474

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 16
• 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: 10
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
2.8986	0.0029	1	2.8581
2.1825	0.1473	50	2.1799
2.1385	0.2947	100	2.1195
2.0831	0.4420	150	2.0977
2.1055	0.5893	200	2.0832
1.9432	0.7366	250	2.0724
2.0102	0.8840	300	2.0634
2.0319	1.0313	350	2.0597
1.8641	1.1786	400	2.0566
2.0955	1.3260	450	2.0536
2.0081	1.4733	500	2.0505
2.0862	1.6206	550	2.0491
1.7684	1.7680	600	2.0481
1.9934	1.9153	650	2.0474

Framework versions

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