INDEX

Explanations

references to firearms or weapons

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

########.

-0.82

 HasFactory

-0.78

fjspx

-0.71

Бахар

-0.69

TestingModule

-0.69

NSIndexPath

-0.67

发表于

-0.65

aarrggbb

-0.63

Дереккөздер

-0.62

 isComment

-0.60

POSITIVE LOGITS

 frame

0.53

 Frame

0.48

パイ

0.47

 pistols

0.45

 handgun

0.45

magem

0.45

 FRAME

0.44

 inox

0.43

 pistol

0.43

洪

0.41

Activations Density 0.053%

references to firearms or weapons

No Comments

No Known Activations