NobodyExistsOnTheInternet/miqu-limarp-70b-logs

See axolotl config

axolotl version: 0.3.0

base_model: 152334H/miqu-1-70b-sf
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: NobodyExistsOnTheInternet/LimaRP
    type: sharegpt
    conversation: chatml
  - path: Doctor-Shotgun/no-robots-sharegpt
    type: sharegpt
    conversation: chatml

chat_template: chatml
dataset_prepared_path: last_run_prepared
val_set_size: 0
output_dir: ./miqu-lora
save_safetensors: true

adapter: qlora
lora_model_dir:

sequence_len: 8192
sample_packing: true

lora_r: 64
lora_alpha: 128
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: 4
micro_batch_size: 1
num_epochs: 4
optimizer: paged_lion_8bit
lr_scheduler: cosine
learning_rate: 0.00025

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
save_total_limit: 2

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
hub_strategy: all_checkpoints
hf_use_auth_token: true

miqu-limarp-70b

This model is a fine-tuned version of 152334H/miqu-1-70b-sf 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: 0.00025
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• total_eval_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 4

Training results

Framework versions

• Transformers 4.37.0.dev0
• Pytorch 2.0.1+cu117
• Datasets 2.16.1
• Tokenizers 0.15.0

Training procedure

The following bitsandbytes quantization config was used during training:

• quant_method: bitsandbytes
• load_in_8bit: False
• load_in_4bit: True
• llm_int8_threshold: 6.0
• llm_int8_skip_modules: None
• llm_int8_enable_fp32_cpu_offload: False
• llm_int8_has_fp16_weight: False
• bnb_4bit_quant_type: nf4
• bnb_4bit_use_double_quant: True
• bnb_4bit_compute_dtype: bfloat16

Framework versions

• PEFT 0.6.0