INDEX

Explanations

elements representing physical structures and their interactions

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gonic

-0.67

CloseOperation

-0.61

ViewInit

-0.60

zonych

-0.59

createCanvas

-0.58

bujete

-0.57

racite

-0.56

 Seidel

-0.56

limsy

-0.56

Hauptartikel

-0.56

POSITIVE LOGITS

 exposed

1.06

 protruding

0.93

 Exposed

0.90

exposed

0.80

 visible

0.78

 protr

0.77

visible

0.76

 exposure

0.74

Exposed

0.74

 overhang

0.74

Activations Density 0.506%

elements representing physical structures and their interactions

No Comments

No Known Activations