INDEX

Explanations

phrases related to copying/pasting content

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

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

DeleteBehavior

-0.58

 MaterialApp

-0.57

 noires

-0.56

 femininas

-0.54

ApiModel

-0.54

gameserver

-0.53

StructEnd

-0.52

SharedCtor

-0.51

atch

-0.51

 Bakan

-0.50

POSITIVE LOGITS

 paste

1.07

 copy

1.04

 Copy

0.99

 clipboard

0.97

 Paste

0.93

 Ctrl

0.91

 COPY

0.89

Copy

0.89

paste

0.88

 drag

0.87

Activations Density 0.398%

phrases related to copying/pasting content

No Comments

No Known Activations