cpt core 4

scripts/cpt_core_model_4.py

@@ -37,17 +37,17 @@ model = FastLanguageModel.get_peft_model(
     r=256, # Choose any number > 0 ! Suggested 8, 16, 32, 64, 128
     # r=16, # Choose any number > 0 ! Suggested 8, 16, 32, 64, 128
     target_modules=[
-        "q_proj", "k_proj", "v_proj", "o_proj",
-        "gate_proj",
-        "up_proj", "down_proj",
-        "embed_tokens", "lm_head",
+        'q_proj', 'k_proj', 'v_proj', 'o_proj',
+        'gate_proj',
+        'up_proj', 'down_proj',
+        'embed_tokens', 'lm_head',
     ],
     lora_alpha=32,
     # lora_alpha=16,
     lora_dropout=0, # Supports any, but = 0 is optimized
     bias='none',    # Supports any, but = "none" is optimized
     # [NEW] "unsloth" uses 30% less VRAM, fits 2x larger batch sizes!
-    # use_gradient_checkpointing="unsloth", # True or "unsloth" for very long context
+    # use_gradient_checkpointing='unsloth', # True or "unsloth" for very long context
     use_gradient_checkpointing=False,
     random_state=23,
     use_rslora=True,    # We support rank stabilized LoRA