INDEX

Explanations

numerical data formatted as version numbers or software specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

fromCharCode

-0.73

getMinutes

-0.62

stdafx

-0.62

'],$

-0.60

labelledby

-0.59

 ffilmiau

-0.59

HttpMethod

-0.59

ITHER

-0.59

Gub

-0.58

 htmlFor

-0.57

POSITIVE LOGITS

subsubsection

0.84

évaluateur

0.69

 tranquille

0.63

 transfieras

0.58

 rodeado

0.57

currentColor

0.56

 ſch

0.56

viders

0.55

vece

0.55

paž

0.54

Activations Density 0.119%

numerical data formatted as version numbers or software specifications

No Comments

No Known Activations