INDEX

Explanations

mentions of magazines

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

amentals

-0.73

 Drit

-0.73

xase

-0.71

EqualsAnd

-0.67

훈

-0.62

 "','

-0.62

 Réponses

-0.61

audiovisuel

-0.59

Kanpo

-0.58

 cristi

-0.58

POSITIVE LOGITS

 magazine

2.08

 magazines

2.01

 MAGAZINE

1.98

 Magazines

1.91

 Magazine

1.88

 mags

1.86

Magazine

1.81

MAGAZINE

1.79

mag

1.77

magazine

1.71

Activations Density 0.063%

mentions of magazines

No Comments

No Known Activations