INDEX

Explanations

phrases indicating presence, existence, or occurrence

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.41.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

GEBURTSDATUM

-0.45

loé

-0.40

ukone

-0.39

Tikang

-0.38

ærk

-0.38

 оригіналу

-0.37

querque

-0.36

KommentareTeilen

-0.36

thello

-0.35

árs

-0.35

POSITIVE LOGITS

est

0.90

 était

0.71

fut

0.66

 doit

0.62

 sera

0.60

 serait

0.60

 devait

0.59

 restera

0.56

 existe

0.55

 vient

0.55

Activations Density 0.215%

phrases indicating presence, existence, or occurrence

No Known Activations