use new modaic nomenclature
This commit is contained in:
20
program.json
20
program.json
@@ -28,7 +28,7 @@
|
|||||||
"finetuning_model": null,
|
"finetuning_model": null,
|
||||||
"launch_kwargs": {},
|
"launch_kwargs": {},
|
||||||
"train_kwargs": {},
|
"train_kwargs": {},
|
||||||
"temperature": null,
|
"temperature": 0,
|
||||||
"max_tokens": 512
|
"max_tokens": 512
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
@@ -61,7 +61,7 @@
|
|||||||
"finetuning_model": null,
|
"finetuning_model": null,
|
||||||
"launch_kwargs": {},
|
"launch_kwargs": {},
|
||||||
"train_kwargs": {},
|
"train_kwargs": {},
|
||||||
"temperature": null,
|
"temperature": 0,
|
||||||
"max_tokens": 512
|
"max_tokens": 512
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
@@ -94,7 +94,7 @@
|
|||||||
"finetuning_model": null,
|
"finetuning_model": null,
|
||||||
"launch_kwargs": {},
|
"launch_kwargs": {},
|
||||||
"train_kwargs": {},
|
"train_kwargs": {},
|
||||||
"temperature": null,
|
"temperature": 0,
|
||||||
"max_tokens": 512
|
"max_tokens": 512
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
@@ -127,7 +127,7 @@
|
|||||||
"finetuning_model": null,
|
"finetuning_model": null,
|
||||||
"launch_kwargs": {},
|
"launch_kwargs": {},
|
||||||
"train_kwargs": {},
|
"train_kwargs": {},
|
||||||
"temperature": null,
|
"temperature": 0,
|
||||||
"max_tokens": 512
|
"max_tokens": 512
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
@@ -160,7 +160,7 @@
|
|||||||
"finetuning_model": null,
|
"finetuning_model": null,
|
||||||
"launch_kwargs": {},
|
"launch_kwargs": {},
|
||||||
"train_kwargs": {},
|
"train_kwargs": {},
|
||||||
"temperature": null,
|
"temperature": 0,
|
||||||
"max_tokens": 512
|
"max_tokens": 512
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
@@ -197,7 +197,7 @@
|
|||||||
"finetuning_model": null,
|
"finetuning_model": null,
|
||||||
"launch_kwargs": {},
|
"launch_kwargs": {},
|
||||||
"train_kwargs": {},
|
"train_kwargs": {},
|
||||||
"temperature": null,
|
"temperature": 0,
|
||||||
"max_tokens": 512
|
"max_tokens": 512
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
@@ -234,7 +234,7 @@
|
|||||||
"finetuning_model": null,
|
"finetuning_model": null,
|
||||||
"launch_kwargs": {},
|
"launch_kwargs": {},
|
||||||
"train_kwargs": {},
|
"train_kwargs": {},
|
||||||
"temperature": null,
|
"temperature": 0,
|
||||||
"max_tokens": 512
|
"max_tokens": 512
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
@@ -271,7 +271,7 @@
|
|||||||
"finetuning_model": null,
|
"finetuning_model": null,
|
||||||
"launch_kwargs": {},
|
"launch_kwargs": {},
|
||||||
"train_kwargs": {},
|
"train_kwargs": {},
|
||||||
"temperature": null,
|
"temperature": 0,
|
||||||
"max_tokens": 512
|
"max_tokens": 512
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
@@ -308,7 +308,7 @@
|
|||||||
"finetuning_model": null,
|
"finetuning_model": null,
|
||||||
"launch_kwargs": {},
|
"launch_kwargs": {},
|
||||||
"train_kwargs": {},
|
"train_kwargs": {},
|
||||||
"temperature": null,
|
"temperature": 0,
|
||||||
"max_tokens": 512
|
"max_tokens": 512
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
@@ -345,7 +345,7 @@
|
|||||||
"finetuning_model": null,
|
"finetuning_model": null,
|
||||||
"launch_kwargs": {},
|
"launch_kwargs": {},
|
||||||
"train_kwargs": {},
|
"train_kwargs": {},
|
||||||
"temperature": null,
|
"temperature": 0,
|
||||||
"max_tokens": 512
|
"max_tokens": 512
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
|
|||||||
@@ -63,14 +63,18 @@ class AttackProgram(PrecompiledProgram):
|
|||||||
**kwargs,
|
**kwargs,
|
||||||
):
|
):
|
||||||
super().__init__(config, **kwargs)
|
super().__init__(config, **kwargs)
|
||||||
attack_model = dspy.LM(model=config.lm, max_tokens=config.max_attack_tokens)
|
|
||||||
|
attack_model = dspy.LM(model=config.lm, max_tokens=config.max_attack_tokens, temperature=config.temperature)
|
||||||
self.get_response = get_response
|
self.get_response = get_response
|
||||||
self.layers = config.num_layers
|
self.layers = config.num_layers
|
||||||
|
|
||||||
self.try_attacks = [dspy.Predict(Attack) for _ in range(self.layers)]
|
self.try_attacks = [dspy.Predict(Attack) for _ in range(self.layers)]
|
||||||
self.critique_attacks = [dspy.Predict(Refine) for _ in range(self.layers)]
|
self.critique_attacks = [dspy.Predict(Refine) for _ in range(self.layers)]
|
||||||
|
|
||||||
self.target_model_name = config.target_lm
|
self.target_model_name = config.target_lm
|
||||||
self.max_attack_tokens = config.max_attack_tokens
|
self.max_attack_tokens = config.max_attack_tokens
|
||||||
self.temperature = config.temperature
|
self.temperature = config.temperature
|
||||||
|
|
||||||
self.set_lm(attack_model)
|
self.set_lm(attack_model)
|
||||||
|
|
||||||
def forward(self, harmful_intent, critique=""):
|
def forward(self, harmful_intent, critique=""):
|
||||||
|
|||||||
Reference in New Issue
Block a user