INDEX

Explanations

names of various taxonomic classifications or biological terms

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

words ending in -an/-ian/-ean

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

 réservé

-0.45

 Wikiseite

-0.45

 cooperación

-0.42

erçe

-0.42

 jadx

-0.42

 TAMBIÉN

-0.42

katapos

-0.41

menistan

-0.41

 Czechoslovakia

-0.41

amitié

-0.41

POSITIVE LOGITS

neurial

0.66

 Rican

0.66

liferay

0.65

costal

0.65

 متعلقه

0.60

 дописавши

0.59

onaceous

0.57

 polymorphic

0.56

 cortical

0.55

 ocular

0.54

Activations Density 0.133%

names of various taxonomic classifications or biological terms

words ending in -an/-ian/-ean

No Comments

No Known Activations