INDEX

Explanations

names and terms associated with specific scientific concepts 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 @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

tory

-0.93

triangleq

-0.78

gory

-0.77

spacer

-0.73

 Woodbury

-0.71

 Tsche

-0.71

tale

-0.71

Референце

-0.68

yssey

-0.68

adecimal

-0.68

POSITIVE LOGITS

IN

1.27

in

1.25

nnnn

1.15

Zin

1.11

nin

1.09

vin

1.09

inn

1.08

HIN

1.08

nnn

1.06

pin

1.05

Activations Density 0.587%

names and terms associated with specific scientific concepts or entities

No Comments

No Known Activations