INDEX

Explanations

references to marine-related topics or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 clots

-1.00

 Réponses

-0.95

ValueStyle

-0.89

 Roskov

-0.88

 Czechoslovakia

-0.87

 engraçadas

-0.86

rhosis

-0.85

 AssemblyCulture

-0.84

 Picchu

-0.84

 Yugos

-0.82

POSITIVE LOGITS

 marine

1.20

 Marine

1.17

Marine

1.03

 marines

0.91

 MARINE

0.91

 Marines

0.88

marine

0.81

 Marin

0.73

${{

0.72

<i>

0.71

Activations Density 0.011%

references to marine-related topics or entities

No Comments

No Known Activations