INDEX

Explanations

expressions of refusal or denial

oai_token-act-pair · gpt-4o-mini Triggered by @bot

"Refuse" / "refused" followed by a verb

np_acts-logits-general · gemini-2.0-flash

refuse to

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SUCCEEDED

-0.54

::$_

-0.52

 tikai

-0.51

]!='

-0.49

boll

-0.47

Tracce

-0.47

CAPTION

-0.46

Subset

-0.45

 selamat

-0.45

feinander

-0.44

POSITIVE LOGITS

AddTagHelper

1.01

 altogether

0.93

any

0.85

 catég

0.81

InjectAttribute

0.79

UnknownFields

0.76

GraphicsUnit

0.72

SourceChecksum

0.71

ьаж

0.70

 ivelany

0.69

Activations Density 0.344%

expressions of refusal or denial

"Refuse" / "refused" followed by a verb

refuse to

No Comments

No Known Activations