INDEX

Explanations

references to people, entities, or terminologies related to technical or academic subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

illingham

-0.47

 espar

-0.45

didReceive

-0.44

 uintptr

-0.43

ast

-0.42

ッズ

-0.42

ständen

-0.41

hasiswa

-0.40

ář

-0.38

шин

-0.38

POSITIVE LOGITS

RegressionTest

0.90

findpost

0.85

ScopeManager

0.80

 myſelf

0.70

InSection

0.68

bewerken

0.65

出版年

0.64

 виправивши

0.64

Démographie

0.64

 gyhoeddwyd

0.63

Activations Density 0.199%

references to people, entities, or terminologies related to technical or academic subjects

No Comments

No Known Activations