INDEX

Explanations

symbolic mathematical expressions and notations

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

 CURIAM

-0.45

 remp

-0.44

onsors

-0.42

nonatomic

-0.42

 esternos

-0.42

 Statt

-0.41

ggars

-0.41

xampp

-0.41

 vers

-0.40

rån

-0.40

POSITIVE LOGITS

تقاوى

0.81

发表于

0.71

saraba

0.70

CppMethod

0.64

 समीक्षाओं

0.63

intenant

0.63

berdayakan

0.63

featureID

0.62

SourceChecksum

0.61

 diary

0.60

Activations Density 1.133%

symbolic mathematical expressions and notations

No Comments

No Known Activations