See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: unsloth/SmolLM-360M-Instruct
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 5a69eb8dcba97959_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/5a69eb8dcba97959_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/6a454368-55f6-4471-aec6-5579150e1b24
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/5a69eb8dcba97959_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: f791392d-4fbb-413b-8436-7f3a97a43fb5
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: f791392d-4fbb-413b-8436-7f3a97a43fb5
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

6a454368-55f6-4471-aec6-5579150e1b24

This model is a fine-tuned version of unsloth/SmolLM-360M-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.8379

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
3.3599	0.0014	1	3.7927
3.656	0.0138	10	3.6697
3.0074	0.0276	20	3.2149
3.2849	0.0414	30	3.0086
2.7319	0.0552	40	2.9370
3.0431	0.0689	50	2.8915
3.1456	0.0827	60	2.8677
2.1565	0.0965	70	2.8533
3.2492	0.1103	80	2.8417
3.0706	0.1241	90	2.8379
2.1927	0.1379	100	2.8379

Framework versions

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