[optim] add support to APOLLO (#6617)

Former-commit-id: 5a252e5a458457adbd19da3b68a3897ad2962824

examples/extras/apollo/llama3_full_sft.yaml

@@ -0,0 +1,45 @@
+### model
+model_name_or_path: meta-llama/Meta-Llama-3-8B-Instruct
+trust_remote_code: true
+
+### method
+stage: sft
+do_train: true
+finetuning_type: full
+use_apollo: true
+apollo_layerwise: true
+apollo_target: mlp,self_attn
+apollo_rank: 128
+apollo_scale: 32.0
+apollo_scale_type: channel
+
+### dataset
+dataset: identity,alpaca_en_demo
+template: llama3
+cutoff_len: 2048
+max_samples: 1000
+overwrite_cache: true
+preprocessing_num_workers: 16
+
+### output
+output_dir: saves/llama3-8b/apollo_full-scale32/sft
+logging_steps: 10
+save_steps: 500
+plot_loss: true
+overwrite_output_dir: true
+
+### train
+per_device_train_batch_size: 1
+gradient_accumulation_steps: 1
+learning_rate: 1.0e-5
+num_train_epochs: 3.0
+lr_scheduler_type: cosine
+warmup_ratio: 0.1
+pure_bf16: true
+ddp_timeout: 180000000
+
+### eval
+val_size: 0.1
+per_device_eval_batch_size: 1
+eval_strategy: steps
+eval_steps: 500