See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: fxmarty/tiny-dummy-qwen2
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - f1d35c010684b293_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/f1d35c010684b293_train_data.json
  type:
    field_input: serialized_infobox
    field_instruction: infobox
    field_output: summary
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 4
flash_attention: false
fp16: true
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: ardaspear/8d3b9a2d-4c9c-4052-9a18-0ee6e5d4c02b
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: 0
logging_steps: 3
lora_alpha: 128
lora_dropout: 0.1
lora_fan_in_fan_out: true
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_memory:
  0: 72GB
max_steps: 50
micro_batch_size: 4
mlflow_experiment_name: /tmp/f1d35c010684b293_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: false
sample_packing: false
saves_per_epoch: 4
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: leixa-personal
wandb_mode: online
wandb_name: 8d3b9a2d-4c9c-4052-9a18-0ee6e5d4c02b
wandb_project: Gradients-On-Two
wandb_run: your_name
wandb_runid: 8d3b9a2d-4c9c-4052-9a18-0ee6e5d4c02b
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

8d3b9a2d-4c9c-4052-9a18-0ee6e5d4c02b

This model is a fine-tuned version of fxmarty/tiny-dummy-qwen2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 11.8981

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• 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
• training_steps: 50

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0180	1	11.9301
11.9306	0.0899	5	11.9296
11.9286	0.1798	10	11.9273
11.9244	0.2697	15	11.9228
11.9214	0.3596	20	11.9166
11.9144	0.4494	25	11.9099
11.9055	0.5393	30	11.9043
11.904	0.6292	35	11.9007
11.8995	0.7191	40	11.8988
11.899	0.8090	45	11.8982
11.8982	0.8989	50	11.8981

Framework versions

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