INDEX

Explanations

numerical values and identifiers related to structured data or entities

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

Appears before a number

np_acts-logits-general · gemini-2.0-flash

James MESSER, Jr

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

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

-0.58

-0.56

-0.55

Kilder

-0.54

-0.53

-0.51

…

-0.50

POSITIVE LOGITS

Personensuche

0.75

tinyos

0.64

estacks

0.61

')}}">

0.60

>({

0.59

rillation

0.59

ophageal

0.59

 StringVar

0.59

addGap

0.58

)"),

0.58

Activations Density 0.013%

numerical values and identifiers related to structured data or entities

Appears before a number

James MESSER, Jr

No Comments

No Known Activations