qwen-2.5-72b-ft-agent

<!-- 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/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>

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

axolotl version: 0.8.0.dev0

adapter: lora
base_model: Qwen/Qwen2.5-72B-Instruct
bf16: auto
dataset_processes: 32
datasets:
  - path: alpaca_dataset.jsonl
    type: alpaca
    system_prompt: system_prompt.txt
    message_property_mappings:
      instruction: instruction
      input: input
      output: output
gradient_accumulation_steps: 4 # Increased to compensate for smaller batch size
gradient_checkpointing: true # Enabled to save VRAM
learning_rate: 0.0002
lisa_layers_attribute: model.layers
load_best_model_at_end: false
load_in_4bit: true # Changed to 4-bit quantization for better memory efficiency
load_in_8bit: false # Disabled 8-bit since we're using 4-bit
lora_alpha: 16
lora_dropout: 0.05
lora_r: 8
lora_target_modules:
  - q_proj
  - v_proj
  - k_proj
  - o_proj
  - gate_proj
  - down_proj
  - up_proj
loraplus_lr_embedding: 1.0e-06
lr_scheduler: cosine
max_prompt_len: 512
mean_resizing_embeddings: false
micro_batch_size: 4 # Reduced from 16 to save memory
num_epochs: 1.0
optimizer: adamw_bnb_8bit
output_dir: ./outputs/mymodel
pretrain_multipack_attn: true
pretrain_multipack_buffer_size: 10000
qlora_sharded_model_loading: false
ray_num_workers: 1
resources_per_worker:
  GPU: 1
sample_packing_bin_size: 200
sample_packing_group_size: 100000
save_only_model: false
save_safetensors: true
sequence_len: 2048 # Reduced from 4096 to save memory
shuffle_merged_datasets: true
skip_prepare_dataset: false
strict: false
train_on_inputs: false
trl:
  log_completions: false
  ref_model_mixup_alpha: 0.9
  ref_model_sync_steps: 64
  sync_ref_model: false
  use_vllm: false
  vllm_device: auto
  vllm_dtype: auto
  vllm_gpu_memory_utilization: 0.9
use_ray: false
val_set_size: 0.0
weight_decay: 0.0

</details><br>

outputs/mymodel

This model is a fine-tuned version of Qwen/Qwen2.5-72B-Instruct on the alpaca_dataset.jsonl 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: 0.0002
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 1.0

Training results

Framework versions

• PEFT 0.14.0
• Transformers 4.49.0
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0