chore: import upstream snapshot with attribution

2026-07-13 13:34:58 +08:00
commit a203934033
1368 changed files with 175001 additions and 0 deletions
@@ -0,0 +1,22 @@
+# loss_scale all to train all tokens
+# use loss_type loss_scale
+# This is just an example
+CUDA_VISIBLE_DEVICES=0 \
+swift sft \
+    --model Qwen/Qwen2.5-7B-Instruct \
+    --tuner_type lora \
+    --dataset 'swift/self-cognition#1000' \
+    --num_train_epochs 1 \
+    --per_device_train_batch_size 1 \
+    --learning_rate 1e-4 \
+    --lora_rank 8 \
+    --lora_alpha 32 \
+    --gradient_accumulation_steps 16 \
+    --eval_steps 100 \
+    --save_steps 100 \
+    --save_total_limit 2 \
+    --logging_steps 5 \
+    --model_author swift \
+    --model_name swift-robot \
+    --loss_scale all \
+    --loss_type loss_scale
@@ -0,0 +1,22 @@
+# `--tuner_type dummy`
+CUDA_VISIBLE_DEVICES=0 \
+swift sft \
+    --model LLM-Research/Phi-4-multimodal-instruct \
+    --dataset 'AI-ModelScope/LaTeX_OCR:human_handwrite#20000' \
+    --load_from_cache_file true \
+    --split_dataset_ratio 0.01 \
+    --tuner_type dummy \
+    --torch_dtype bfloat16 \
+    --num_train_epochs 1 \
+    --per_device_train_batch_size 1 \
+    --per_device_eval_batch_size 1 \
+    --learning_rate 1e-4 \
+    --gradient_accumulation_steps 16 \
+    --eval_steps 200 \
+    --save_steps 200 \
+    --save_total_limit 2 \
+    --logging_steps 5 \
+    --max_length 2048 \
+    --output_dir output \
+    --warmup_ratio 0.05 \
+    --dataloader_num_workers 4