T3Q-LLM-sft1.0-dpo1.0_100QA_10epochs

<!-- 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: T3Q-LLM/T3Q-LLM-sft1.0-dpo1.0
base_model_config: T3Q-LLM/T3Q-LLM-sft1.0-dpo1.0
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
is_llama_derived_model: true
hub_model_id: T3Q-LLM-sft1.0-dpo1.0_100QA

load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: admin_data.csv
    type: alpaca
      # The below are defaults. only set what's needed if you use a different column name.
      # system_prompt: ""
      # system_format: "{system}"
      # field_system: system
      # field_instruction: instruction
      # field_input: input
      # field_output: output

      # format: |-
      #   Human: {instruction} {input}
      #   Assistant:

      # no_input_format: "{instruction} "

# dataset_prepared_path: yanolja_preprocessed_data
dataset_prepared_path: last_run_prepared
val_set_size: 0.2
output_dir: ./T3Q-LLM-sft1.0-dpo1.0_100QA

adapter: qlora
lora_model_dir: 

# device_map: [0,1,3]

sequence_len: 4096
sample_packing: false

lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_modules:
lora_target_linear: true
lora_fan_in_fan_out: 

wandb_project: axolotl_T3Q
wandb_entity: 
wandb_watch: 
wandb_run_id: T3Q_mod_100
wandb_log_model: 

gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 10
optimizer: paged_adamw_32bit
lr_scheduler: cosine
learning_rate: 0.0002

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

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

warmup_steps: 100
eval_steps: 0.01
save_strategy: epoch
save_steps: 
debug: 
deepspeed: 
weight_decay: 0.0
fsdp: 
fsdp_config: 
special_tokens:
    bos_token: "<s>"
    eos_token: "<|im_end|>"
    unk_token: "<unk>"
    pad_token: "</s>"  # EOS와 PAD가 동일

</details><br>

T3Q-LLM-sft1.0-dpo1.0_100QA

This model is a fine-tuned version of T3Q-LLM/T3Q-LLM-sft1.0-dpo1.0 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6629

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: 0.0002
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• total_eval_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 10

Training results

Framework versions

• PEFT 0.10.0
• Transformers 4.40.1
• Pytorch 2.1.2+cu121
• Datasets 2.15.0
• Tokenizers 0.19.1