INDEX

Explanations

phrases related to achieving or validating tasks and objectives

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

 occaf

-0.62

 pleaſure

-0.61

 raiſ

-0.57

 beraber

-0.53

 ſel

-0.52

 houſe

-0.52

 caufe

-0.52

 perſon

-0.50

 cauſe

-0.50

 uſed

-0.50

POSITIVE LOGITS

RectangleBorder

0.68

ScopeManager

0.62

MLLoader

0.60

 위해

0.57

ůli

0.56

phrine

0.55

 समीक्षाओं

0.54

NCT

0.54

="@+

0.53

InSection

0.52

Activations Density 0.548%

phrases related to achieving or validating tasks and objectives

No Comments

No Known Activations