See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: fxmarty/tiny-llama-fast-tokenizer
bf16: true
chat_template: llama3
data_processes: 24
dataset_prepared_path: null
datasets:
- data_files:
  - d683d2cbbbb650a8_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/d683d2cbbbb650a8_train_data.json
  type:
    field_input: Ingredientes
    field_instruction: Nombre
    field_output: Pasos
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map: auto
do_eval: true
early_stopping_patience: 4
eval_batch_size: 4
eval_max_new_tokens: 128
eval_steps: 50
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: true
hub_model_id: cimol/3815d577-e401-4bcc-9b7f-fff1f2f8bac6
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 7.0e-05
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 128
lora_dropout: 0.04
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
lr_scheduler_warmup_steps: 50
max_grad_norm: 1.0
max_memory:
  0: 75GB
max_steps: 200
micro_batch_size: 8
mlflow_experiment_name: /tmp/d683d2cbbbb650a8_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optim_args:
  adam_beta1: 0.9
  adam_beta2: 0.95
  adam_epsilon: 1e-8
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: 50
saves_per_epoch: null
seed: 17333
sequence_len: 1024
special_tokens:
  pad_token: </s>
strict: false
tf32: true
tokenizer_type: AutoTokenizer
total_train_batch_size: 32
train_batch_size: 8
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: eaf0ed6f-f6ec-417e-9ead-2af5434e5974
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: eaf0ed6f-f6ec-417e-9ead-2af5434e5974
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

3815d577-e401-4bcc-9b7f-fff1f2f8bac6

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

• Loss: 10.3459

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: 7e-05
• train_batch_size: 8
• eval_batch_size: 4
• seed: 17333
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-8
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• training_steps: 200

Training results

Training Loss	Epoch	Step	Validation Loss
10.3761	0.0017	1	10.3765
10.3542	0.0847	50	10.3515
10.3469	0.1693	100	10.3472
10.348	0.2540	150	10.3461
10.3449	0.3387	200	10.3459

Framework versions

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