adapter: lora
base_model: OpenAssistant/oasst-sft-4-pythia-12b-epoch-3.5
bf16: auto
dataset_prepared_path: null
datasets:
- data_files:
  - 296cd3b6a996ecf6_train_data.json
  ds_type: json
  format: custom
  path: 296cd3b6a996ecf6_train_data.json
  type:
    field: null
    field_input: null
    field_instruction: user
    field_output: chip2
    field_system: null
    format: null
    no_input_format: null
    system_format: '{system}'
    system_prompt: ''
early_stopping_patience: null
evals_per_epoch: 2
gradient_accumulation_steps: 1
group_by_length: false
hub_model_id: taopanda-1/17ba6118-af07-459a-959a-678a189f44a4
learning_rate: 1.0e-05
load_in_8bit: true
local_rank: null
logging_steps: 1
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: true
lora_model_dir: null
lora_r: 16
lora_target_linear: null
lora_target_modules:
- query_key_value
micro_batch_size: 4
num_epochs: 1
output_dir: ./outputs/lora-alpaca-pythia/taopanda-1_03508cf1-5aff-47a6-b375-7a4ea203b6a0
resume_from_checkpoint: null
seed: 13400
sequence_len: 512
tf32: true
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: fatcat87-taopanda
wandb_log_model: null
wandb_mode: online
wandb_name: taopanda-1_03508cf1-5aff-47a6-b375-7a4ea203b6a0
wandb_project: subnet56
wandb_runid: taopanda-1_03508cf1-5aff-47a6-b375-7a4ea203b6a0
wandb_watch: null
weight_decay: 0.1

Visualize in Weights & Biases

17ba6118-af07-459a-959a-678a189f44a4

This model is a fine-tuned version of OpenAssistant/oasst-sft-4-pythia-12b-epoch-3.5 on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.2511

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: 1e-05
train_batch_size: 4
eval_batch_size: 4
seed: 13400
distributed_type: multi-GPU
num_devices: 4
total_train_batch_size: 16
total_eval_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 100
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
2.1448	0.0001	1	1.8297
1.1969	0.5	6237	1.2616
0.922	1.0	12474	1.2511

Framework versions

PEFT 0.11.1
Transformers 4.42.3
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.19.1