See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: fxmarty/tiny-llama-fast-tokenizer
bf16: false
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 684f8537349c76ab_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/684f8537349c76ab_train_data.json
  type:
    field_input: system
    field_instruction: user
    field_output: assistant
    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: 4
gradient_checkpointing: false
group_by_length: false
hub_model_id: lesso07/e6bdfa97-5090-445b-9017-ba45d2da0a32
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
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: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine
max_memory:
  0: 70GiB
max_steps: 50
micro_batch_size: 2
mlflow_experiment_name: /tmp/684f8537349c76ab_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 25
save_strategy: steps
sequence_len: 1024
special_tokens:
  pad_token: </s>
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: e6bdfa97-5090-445b-9017-ba45d2da0a32
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: e6bdfa97-5090-445b-9017-ba45d2da0a32
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

e6bdfa97-5090-445b-9017-ba45d2da0a32

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: 10.3802

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_TORCH 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
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
10.3786	0.0001	1	10.3823
10.3817	0.0007	5	10.3822
10.385	0.0013	10	10.3819
10.3786	0.0020	15	10.3816
10.3836	0.0026	20	10.3812
10.3814	0.0033	25	10.3808
10.3842	0.0039	30	10.3805
10.3843	0.0046	35	10.3803
10.3792	0.0052	40	10.3802
10.3768	0.0059	45	10.3802
10.3826	0.0065	50	10.3802

Framework versions

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