INDEX

Explanations

references to guidance or instructions on how to begin a task

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tagHelperRunner

-0.63

 onPostExecute

-0.50

twimg

-0.45

Autoritní

-0.44

 autorytatywna

-0.43

wix

-0.42

CheckBreak

-0.41

ſelves

-0.40

TokenNameRBRACE

-0.40

 Exactos

-0.40

POSITIVE LOGITS

 start

2.56

 starts

2.44

 starting

2.38

 begin

2.30

 beginning

2.27

 begins

2.22

 started

2.20

 Start

2.13

 시작

2.09

start

2.08

Activations Density 0.989%

references to guidance or instructions on how to begin a task

No Comments

No Known Activations