a1907fd4fe* feat(studio): add DoRA support to studio * fix: added use_dora fast encoder LoraConfig and gated use_dora on AdapterMethod * fix(studio) serverside normalization for use_dora=true - add note documenting use_dora is silently dropped on diffusion * fix: dora button disabled on mac, add preflight guard on GGUF lora export, mismatch now correctly falls through to existing error instead of silently no-opping * Studio: add dora to the WizardState LoRA variant union for consistency * Reject --use_dora on the MLX (Apple Silicon) CLI path --------- Co-authored-by: danielhanchen <unslothai@gmail.com>
43 lines
940 B
YAML
43 lines
940 B
YAML
model: unsloth/Qwen2-VL-2B-Instruct-bnb-4bit
|
|
|
|
data:
|
|
dataset: philschmid/amazon-product-descriptions-vlm
|
|
format_type: auto
|
|
|
|
training:
|
|
training_type: lora
|
|
max_seq_length: 2048
|
|
load_in_4bit: true
|
|
output_dir: outputs
|
|
num_epochs: 1
|
|
learning_rate: 0.0002
|
|
batch_size: 1
|
|
gradient_accumulation_steps: 4
|
|
warmup_steps: 5
|
|
max_steps: 0
|
|
save_steps: 0
|
|
weight_decay: 0.001
|
|
random_seed: 3407
|
|
packing: false
|
|
train_on_completions: false
|
|
gradient_checkpointing: "unsloth"
|
|
|
|
lora:
|
|
lora_r: 64
|
|
lora_alpha: 16
|
|
lora_dropout: 0.0
|
|
target_modules: "" # vision uses vision_all_linear by default
|
|
vision_all_linear: true
|
|
use_rslora: false
|
|
use_loftq: false
|
|
use_dora: false
|
|
finetune_vision_layers: true
|
|
finetune_language_layers: true
|
|
finetune_attention_modules: true
|
|
finetune_mlp_modules: true
|
|
|
|
logging:
|
|
enable_wandb: false
|
|
wandb_project: unsloth-training
|
|
enable_tensorboard: false
|
|
tensorboard_dir: runs
|