See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: EleutherAI/gpt-neo-1.3B
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - c8397f45edf002b3_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/c8397f45edf002b3_train_data.json
  type:
    field_input: input
    field_instruction: instruction
    field_output: output
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
device_map:
  ? ''
  : 0,1,2,3,4,5,6,7
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: false
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Alphatao/633e6a49-81e3-4492-8a1e-32975c89d1a4
hub_repo: null
hub_strategy: null
hub_token: null
learning_rate: 0.0002
load_best_model_at_end: true
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
lora_target_modules: null
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 2652
micro_batch_size: 4
mlflow_experiment_name: /tmp/c8397f45edf002b3_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
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
special_tokens:
  pad_token: <|endoftext|>
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.04
wandb_entity: null
wandb_mode: online
wandb_name: 7bee0287-1da9-4e6d-9b18-0f8342b3f344
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 7bee0287-1da9-4e6d-9b18-0f8342b3f344
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

633e6a49-81e3-4492-8a1e-32975c89d1a4

This model is a fine-tuned version of EleutherAI/gpt-neo-1.3B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.6668

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: 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: 2652

Training results

Training Loss	Epoch	Step	Validation Loss
28.7921	0.0007	1	3.6046
19.3797	0.0654	100	2.4536
16.8527	0.1308	200	2.2399
18.0205	0.1962	300	2.1347
16.2049	0.2616	400	2.0673
15.984	0.3270	500	2.0087
15.9218	0.3924	600	1.9615
15.4433	0.4579	700	1.9172
16.0857	0.5233	800	1.8853
15.2135	0.5887	900	1.8540
13.9709	0.6541	1000	1.8281
14.4289	0.7195	1100	1.8007
13.6533	0.7849	1200	1.7801
13.114	0.8503	1300	1.7613
13.2745	0.9157	1400	1.7429
13.6816	0.9811	1500	1.7293
13.9706	1.0470	1600	1.7164
13.6014	1.1124	1700	1.7073
13.8166	1.1778	1800	1.6967
14.0799	1.2432	1900	1.6885
14.2287	1.3086	2000	1.6831
13.6417	1.3740	2100	1.6771
14.0925	1.4395	2200	1.6727
12.2913	1.5049	2300	1.6698
12.4135	1.5703	2400	1.6679
13.7286	1.6357	2500	1.6669
12.4977	1.7011	2600	1.6668

Framework versions

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