ebbd7fa847
* fix: clarify incompat * fix: transformers api change upstream * fix: add pre prop * feat: add examples * chore: cleanup * chore: update readme
57 lines
1017 B
YAML
57 lines
1017 B
YAML
base_model: axolotl-ai-co/Mistral-Medium-3.5-128B-BF16
|
|
|
|
plugins:
|
|
- axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
|
|
|
|
load_in_4bit: true
|
|
|
|
datasets:
|
|
- path: fozziethebeat/alpaca_messages_2k_test
|
|
type: chat_template
|
|
|
|
dataset_prepared_path: last_run_prepared
|
|
val_set_size: 0.1
|
|
output_dir:
|
|
|
|
adapter: qlora
|
|
lora_model_dir:
|
|
|
|
sequence_len: 2048
|
|
sample_packing: true
|
|
|
|
lora_r: 32
|
|
lora_alpha: 16
|
|
lora_dropout: 0
|
|
# prevents targeting vision layers
|
|
lora_target_modules: 'model.language_model.layers.[\d]+.(mlp|self_attn).(up|down|gate|q|k|v|o)_proj'
|
|
|
|
lora_mlp_kernel: true
|
|
lora_qkv_kernel: true
|
|
lora_o_kernel: true
|
|
|
|
wandb_project:
|
|
wandb_entity:
|
|
wandb_watch:
|
|
wandb_name:
|
|
wandb_log_model:
|
|
|
|
gradient_accumulation_steps: 4
|
|
micro_batch_size: 2
|
|
num_epochs: 1
|
|
max_steps: 10
|
|
optimizer: adamw_bnb_8bit
|
|
lr_scheduler: cosine
|
|
learning_rate: 0.0002
|
|
|
|
bf16: auto
|
|
tf32: false
|
|
|
|
gradient_checkpointing: true
|
|
resume_from_checkpoint:
|
|
logging_steps: 1
|
|
flash_attention: true
|
|
|
|
warmup_ratio: 0.1
|
|
evals_per_epoch: 1
|
|
saves_per_epoch: 1
|