ConicCat/Qwen-DGM-32B-V1

See axolotl config

axolotl version: 0.10.0

base_model: Qwen/Qwen2.5-32B-Instruct

plugins:
  - axolotl.integrations.liger.LigerPlugin
  - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin

liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_layer_norm: true
liger_fused_linear_cross_entropy: false

lora_mlp_kernel: true
lora_qkv_kernel: true
lora_o_kernel: true

load_in_8bit: false
load_in_4bit: false

sequence_len: 5120
max_sample_length: 5120

sample_packing: true
gradient_checkpointing: true
flash_attention: true

bf16: true
tf32: true


datasets:
  - path: ConicCat/DGM-Testing
    type: chat_template
    chat_template: chatml
    roles_to_train: []
    message_field_training: train

adapter: lora
lora_r: 64
lora_alpha: 128
lora_dropout: 0.0
lora_target_linear: true
use_tensorboard: true

optimizer: paged_adamw_8bit
learning_rate: 2.5e-5
loraplus_lr_ratio: 16

# Training arguments
output_dir: ./Qwen-DGM-32B
num_epochs: 3
micro_batch_size: 1
gradient_accumulation_steps: 16
warmup_ratio: 0.05
lr_scheduler: 'constant_with_warmup'
max_grad_norm: 1
logging_steps: 1
seed: 42
save_strategy: epoch

Qwen-DGM-32B

This model is a fine-tuned version of Qwen/Qwen2.5-32B-Instruct on the ConicCat/DGM-Testing 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: 2.5e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant_with_warmup
• lr_scheduler_warmup_steps: 20
• training_steps: 408

Training results

Framework versions

• PEFT 0.15.2
• Transformers 4.52.3
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.4