library_name: peft
tags:
- axolotl
- dpo
- trl
- dpo
- generated_from_trainer
base_model: NobodyExistsOnTheInternet/3epoch-miqu-limarp
model-index: - name: miqu-limarp-70b-dpo
results: []
See axolotl config
axolotl version: 0.4.0
base_model: NobodyExistsOnTheInternet/3epoch-miqu-limarp
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
load_in_8bit: false
load_in_4bit: true
strict: false
datasets:
- path: NobodyExistsOnTheInternet/Fixed-FilteredTruthyDPO
split: train
type: chatml.intel
- path: NobodyExistsOnTheInternet/ToxicDPOqa
split: train
type: chatml.intel
- path: NobodyExistsOnTheInternet/Fixed-Distilabel-intel-orca-dpo-pairs
split: train
type: chatml.intel
- path: NobodyExistsOnTheInternet/Fixed-gutenberg-dpo-v0.1
split: train
type: chatml.intel
chat_template: chatml
dataset_prepared_path: last_run_prepared
val_set_size: 0
output_dir: ./miqu-lora
save_safetensors: true
save_steps: 300
rl: dpo
chat_template: chatml
adapter: qlora
lora_model_dir:
sequence_len: 768
lora_r: 32
lora_alpha: 64
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
lora_modules_to_save:
- embed_tokens
- lm_head
wandb_project: miqu-lora
wandb_entity:
wandb_watch:
wandb_run_id:
wandb_log_model:
gradient_accumulation_steps: 2
micro_batch_size: 1
num_epochs: 3
optimizer: paged_lion_8bit
lr_scheduler: cosine
learning_rate: 0.0000014
train_on_inputs: false
group_by_length: false
bf16: true
fp16: false
tf32: false
gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
auto_resume_from_checkpoints: true
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true
warmup_steps: 10
eval_table_size:
weight_decay: 0
special_tokens:
bos_token: "<s>"
eos_token: "<|im_end|>"
unk_token: "</s>"
tokens:
- "<|im_start|>"
- "<|im_end|>"
neftune_noise_alpha: 5
hub_model_id: NobodyExistsOnTheInternet/miqu-limarp-70b-dpo
hub_strategy: all_checkpoints
hf_use_auth_token: true
push_to_hub: true
rl_adapter_ref_model: false
miqu-limarp-70b-dpo
This model was trained from scratch on the None dataset.
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: 1.4e-06
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- total_eval_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 10
- training_steps: 3960
Training results
Framework versions
- PEFT 0.8.2.dev0
- Transformers 4.37.0
- Pytorch 2.0.1+cu117
- Datasets 2.16.1
- Tokenizers 0.15.0