See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: EleutherAI/gpt-neo-125m
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 74adf53c74422e9c_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/74adf53c74422e9c_train_data.json
  type:
    field_input: material
    field_instruction: questions
    field_output: gpt4_answer
    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: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
gradient_clipping: 1.0
group_by_length: false
hub_model_id: leixa/94fada21-acb7-4c08-9eb5-5b93f29cf10d
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: 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_steps: 100
micro_batch_size: 8
mlflow_experiment_name: /tmp/74adf53c74422e9c_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: null
sample_packing: false
saves_per_epoch: 4
sequence_len: 1024
special_tokens:
  pad_token: <|endoftext|>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: techspear-hub
wandb_mode: online
wandb_name: 31f4ffea-837d-4756-a086-902d589e7281
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 31f4ffea-837d-4756-a086-902d589e7281
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

94fada21-acb7-4c08-9eb5-5b93f29cf10d

This model is a fine-tuned version of EleutherAI/gpt-neo-125m on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.7715

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: 8
• eval_batch_size: 8
• seed: 42
• 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=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• training_steps: 100

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0049	1	3.1936
12.1925	0.0441	9	3.1812
12.2752	0.0881	18	3.0855
12.1308	0.1322	27	2.9765
11.317	0.1763	36	2.9036
11.3475	0.2203	45	2.8526
10.8064	0.2644	54	2.8190
10.92	0.3084	63	2.7968
10.7613	0.3525	72	2.7828
11.153	0.3966	81	2.7752
10.7842	0.4406	90	2.7722
11.1974	0.4847	99	2.7715

Framework versions

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