MiquMaid-v2-70B-DPO-LoRA

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>

<details><summary>See axolotl config</summary>

axolotl version: 0.4.0

base_model: NeverSleep/MiquMaid-v2-70B
model_type: LlamaForCausalLM
tokenizer_type: LlamaTokenizer
is_llama_derived_model: true

load_in_8bit: false
load_in_4bit: true
strict: false

rl: dpo
datasets:
  - path: NobodyExistsOnTheInternet/ToxicDPOqa
    split: train
    type: chatml.alpaca
  - path: Undi95/toxic-dpo-v0.1-NoWarning
    split: train
    type: chatml.alpaca2
dataset_prepared_path:
val_set_size: 0.0
output_dir: ./out

adapter: qlora
lora_model_dir:

sequence_len: 1024
sample_packing: false
pad_to_sequence_len: true

lora_r: 16
lora_alpha: 8
lora_dropout: 0.05
lora_target_modules:
lora_target_linear: true
lora_fan_in_fan_out:

wandb_project: MiquMaidDPO
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 3
num_epochs: 3
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.000001

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

gradient_checkpointing: true
gradient_checkpointing_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:
saves_per_epoch: 1
debug:
deepspeed: ./axolotl/deepspeed_configs/zero2.json
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
  bos_token: "<s>"
  eos_token: "</s>"
  unk_token: "<unk>"

</details><br>

out

This model is a fine-tuned version of NeverSleep/MiquMaid-v2-70B 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: 1e-06
• train_batch_size: 3
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 3
• gradient_accumulation_steps: 4
• total_train_batch_size: 36
• total_eval_batch_size: 24
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• training_steps: 598

Training results

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.16.1
• Tokenizers 0.15.0