base_model: Qwen/Qwen3.5-35B-A3B-Base

plugins:
  - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
  - axolotl.integrations.kernels.KernelsPlugin
  - axolotl.integrations.liger.LigerPlugin
use_kernels: true
use_scattermoe: true
liger_layer_norm: true
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_rms_norm_gated: true

torch_compile: false

chat_template: qwen3_5
datasets:
  - path: mlabonne/FineTome-100k
    type: chat_template
    split: train[:20%]
    field_messages: conversations
    message_property_mappings:
      role: from
      content: value

val_set_size: 0.0
output_dir: ./outputs/out
dataset_prepared_path: last_run_prepared

sequence_len: 2048
sample_packing: true

load_in_4bit: true
quantize_moe_experts: true
adapter: qlora
lora_r: 16
lora_alpha: 32
lora_dropout: 0
lora_target_modules:
  - q_proj
  - k_proj
  - v_proj
  - o_proj
  # Add gate_up_proj and down_proj to also target shared experts (nn.Linear):
  # - gate_up_proj
  # - down_proj

# Target routed experts (3D nn.Parameter tensors, not nn.Linear — use lora_target_parameters):
# lora_target_parameters:
#   - mlp.experts.gate_up_proj
#   - mlp.experts.down_proj

lora_qkv_kernel: true
lora_o_kernel: true
lora_mlp_kernel: false

wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 2
micro_batch_size: 4
num_epochs: 1
optimizer: adamw_torch_8bit
lr_scheduler: cosine
learning_rate: 0.0002

bf16: auto
tf32: true

gradient_checkpointing: true
activation_offloading: true
resume_from_checkpoint:
logging_steps: 1
flash_attention: true

warmup_ratio: 0.1
evals_per_epoch: 4
saves_per_epoch: 1
weight_decay: 0.0
special_tokens: