INDEX

Explanations

references to scientific figures or tables

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

Tokens enclosed in double square brackets

np_acts-logits-general · gemini-2.0-flash

text surrounding brackets

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

Ro

-0.70

 Schmid

-0.60

CA

-0.60

CH

-0.59

Am

-0.59

SE

-0.59

AM

-0.58

lande

-0.58

 Vann

-0.57

AB

-0.57

POSITIVE LOGITS

\]

1.15

↵↵

0.82

 مشين

0.73

 متعلقه

0.66

 servici

0.66

 समीक्षाओं

0.65

 مرئيه

0.63

KommentareTeilen

0.59

 AppColors

0.57

carril

0.57

Activations Density 0.255%

references to scientific figures or tables

Tokens enclosed in double square brackets

text surrounding brackets

No Comments

No Known Activations