See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: EleutherAI/pythia-160m
bf16: true
chat_template: llama3
dataloader_num_workers: 12
dataset_prepared_path: null
datasets:
- data_files:
  - 634f0f12aca7b9a2_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/634f0f12aca7b9a2_train_data.json
  type:
    field_instruction: input
    field_output: output
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map: auto
do_eval: true
early_stopping_patience: 3
eval_batch_size: 8
eval_max_new_tokens: 128
eval_steps: 300
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: nttx/c616b2eb-816e-432d-b1e2-4d58d7d929da
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 3e-5
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 10
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_grad_norm: 1.0
max_memory:
  0: 75GB
max_steps: 3000
micro_batch_size: 8
mlflow_experiment_name: /tmp/634f0f12aca7b9a2_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 100
optim_args:
  adam_beta1: 0.9
  adam_beta2: 0.999
  adam_epsilon: 1e-8
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: 300
saves_per_epoch: null
sequence_len: 1024
special_tokens:
  pad_token: <|endoftext|>
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: c06b0827-5e7d-48ee-8b16-5f37d3ed9abf
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: c06b0827-5e7d-48ee-8b16-5f37d3ed9abf
warmup_steps: 50
weight_decay: 0.1
xformers_attention: null

c616b2eb-816e-432d-b1e2-4d58d7d929da

This model is a fine-tuned version of EleutherAI/pythia-160m on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.8270

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.999,adam_epsilon=1e-8
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 50
• training_steps: 3000

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0013	1	3.3130
12.7621	0.3984	300	3.0850
12.5797	0.7968	600	2.9910
12.177	1.1952	900	2.9410
12.057	1.5936	1200	2.9013
11.5501	1.9920	1500	2.8809
11.8135	2.3904	1800	2.8527
11.6053	2.7888	2100	2.8468
11.4575	3.1873	2400	2.8342
11.4659	3.5857	2700	2.8281
11.9571	3.9841	3000	2.8270

Framework versions

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