Files
axolotl/examples/nemotron-h/120b-a12b-qlora.yaml

83 lines
1.6 KiB
YAML

base_model: nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16
plugins:
- axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
- axolotl.integrations.liger.LigerPlugin
liger_layer_norm: true
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_rms_norm_gated: true
# LoRA kernel patches are incompatible with this architecture — see README.
lora_mlp_kernel: false
lora_qkv_kernel: false
lora_o_kernel: false
chat_template: tokenizer_default
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: 4096
sample_packing: true
load_in_4bit: true
quantize_moe_experts: true
adapter: qlora
lora_r: 16
lora_alpha: 32
lora_dropout: 0.0
lora_target_modules:
- q_proj
- k_proj
- v_proj
- o_proj
# To also train MoE expert weights, add them via lora_target_parameters
# (they are 3D nn.Parameter tensors, not nn.Linear — no gate_proj):
# lora_target_parameters:
# - up_proj
# - down_proj
wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:
gradient_accumulation_steps: 4
micro_batch_size: 1
num_epochs: 1
optimizer: adamw_torch_4bit
lr_scheduler: cosine
learning_rate: 0.0002
bf16: auto
tf32: true
gradient_checkpointing: true
gradient_checkpointing_kwargs:
use_reentrant: false
resume_from_checkpoint:
logging_steps: 1
flash_attention: true
warmup_ratio: 0.1
evals_per_epoch: 2
saves_per_epoch: 1
weight_decay: 0.0
special_tokens: