INDEX

Explanations

imperative verbs and cooking directions

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

now, prepare or cook

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 Мексичка

-0.65

 increí

-0.62

WriteTagHelper

-0.62

pulseira

-0.62

elemField

-0.61

Ӕ

-0.59

 müſſen

-0.59

évaluateur

-0.59

stateProvider

-0.59

jsonwebtoken

-0.57

POSITIVE LOGITS

 مشين

0.38

‍

0.33

 prepare

0.31

UrlEncoded

0.31

 prep

0.31

Preparing

0.30

 Prep

0.30

Prepare

0.30

 Prepare

0.29

prep

0.28

Activations Density 0.004%

imperative verbs and cooking directions

now, prepare or cook

No Comments

No Known Activations