INDEX

Explanations

numerical data and sequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

SOT

-0.56

-0.52

bor

-0.51

-0.47

-0.46

Bor

-0.45

lú

-0.44

Vaj

-0.44

tikzpicture

-0.43

POSITIVE LOGITS

 propOrder

1.01

ajuku

0.94

featureID

0.90

InitVars

0.89

Хьажоргаш

0.86

 دیکھیے

0.85

 وتسجيلات

0.81

تقاوى

0.81

UnsafeEnabled

0.79

 Савезне

0.79

Activations Density 0.167%

numerical data and sequences

No Comments

No Known Activations