Hugging Face's logo

gardner
/
TinyLlama-1.1B-SlimOrca-DPO

PEFT
TensorBoard
Safetensors
English
llama
axolotl
dpo
trl
Generated from Trainer
8-bit precision
bitsandbytes
Model card Files Metrics Community
TinyLlama-1.1B-SlimOrca-DPO / README.md
gardner's picture
gardner
Update README.md
bcf80a4 verified
preview code
|
raw
history blame contribute delete
3.47 kB
metadata
library_name: peft
tags:
  - axolotl
  - dpo
  - trl
  - dpo
  - generated_from_trainer
base_model: gardner/TinyLlama-1.1B-SlimOrca
model-index:
  - name: TinyLlama-1.1B-SlimOrca-DPO
    results: []
license: apache-2.0
datasets:
  - argilla/distilabel-intel-orca-dpo-pairs
language:
  - en

TinyLlama-1.1B-SlimOrca-DPO

This model was trained from gardner/TinyLlama-1.1B-SlimOrca on the argilla/distilabel-intel-orca-dpo-pairs dataset.

image/jpeg

Built with Axolotl

See axolotl config

axolotl version: 0.4.0

base_model: gardner/TinyLlama-1.1B-SlimOrca
model_type: LlamaForCausalLM
tokenizer_type: LlamaTokenizer
chat_template: chatml

is_llama_derived_model: true

load_in_8bit: true
load_in_4bit: false
strict: false

rl: dpo
datasets:
  - path: argilla/distilabel-intel-orca-dpo-pairs
    split: train
    type: chatml.gardner

dataset_prepared_path: ./dsprepare/argilla/distilabel-intel-orca-dpo-pairs
val_set_size: 0.05
output_dir: ./tinyllama-1.1b-slimorca-dpo
hub_model_id: gardner/TinyLlama-1.1B-SlimOrca-DPO

sequence_len: 4096
sample_packing: false
pad_to_sequence_len: false

adapter: lora
lora_model_dir:

lora_r: 256
lora_alpha: 128
lora_dropout: 0.05
lora_target_linear: true
lora_modules_to_save:
lora_fan_in_fan_out:
lora_target_modules:
  - gate_proj
  - down_proj
  - up_proj
  - q_proj
  - v_proj
  - k_proj
  - o_proj


wandb_project: tinyllama
wandb_entity: gardner
wandb_name: tinyllama-distilabel-intel-orca-dpo-pairs

gradient_accumulation_steps: 1
micro_batch_size: 2
num_epochs: 3
optimizer: paged_adamw_8bit
adam_beta2: 0.95
adam_epsilion: 0.00001
lr_scheduler: linear
learning_rate: 1.414e-5

train_on_inputs: false
group_by_length: false
bf16: true
fp16: false
tf32: false

gradient_checkpointing: true
gradient_checkpoint_kwargs:
  use_reentrant: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_steps: 10
eval_steps:
eval_table_size:
eval_table_max_new_tokens: 128
save_steps: 45
debug:
deepspeed:
weight_decay: 0.1
fsdp:
fsdp_config:
special_tokens:
save_safetensors: true

dataloader_num_workers: 16
dataloader_pin_memory: true

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: 1.414e-05
  • train_batch_size: 2
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 10
  • training_steps: 19289

Training results

Framework versions

  • PEFT 0.7.1
  • Transformers 4.37.0
  • Pytorch 2.1.2+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.0