See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: unsloth/SmolLM-1.7B
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - fb22517d47ba5b91_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/fb22517d47ba5b91_train_data.json
  type:
    field_input: qwq
    field_instruction: problem
    field_output: text
    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: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 16
gradient_checkpointing: true
group_by_length: false
hub_model_id: Romain-XV/4157dfa5-3988-47c6-9e3b-00f818ce862e
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/fb22517d47ba5b91_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: 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: 970212e9-dc67-48c8-8ec8-461afe128ee2
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 970212e9-dc67-48c8-8ec8-461afe128ee2
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

4157dfa5-3988-47c6-9e3b-00f818ce862e

This model is a fine-tuned version of unsloth/SmolLM-1.7B on the None dataset. It achieves the following results on the evaluation set:

• Loss: nan

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: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.3936	0.0008	1	nan
0.2524	0.0401	50	nan
0.0903	0.0803	100	nan
0.0332	0.1204	150	nan
0.0204	0.1605	200	nan
0.0032	0.2006	250	nan
0.005	0.2408	300	nan
0.0009	0.2809	350	nan
0.0001	0.3210	400	nan
0.0005	0.3612	450	nan
0.0004	0.4013	500	nan
0.0008	0.4414	550	nan
0.0001	0.4815	600	nan
0.0003	0.5217	650	nan
0.0001	0.5618	700	nan
0.0003	0.6019	750	nan
0.0001	0.6421	800	nan
0.0001	0.6822	850	nan
0.0002	0.7223	900	nan
0.0002	0.7624	950	nan
0.0001	0.8026	1000	nan
0.0001	0.8427	1050	nan
0.0002	0.8828	1100	nan
0.0001	0.9230	1150	nan
0.0025	0.9631	1200	nan

Framework versions

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