INDEX

Explanations

references to pharmaceutical and medical terms

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 betweenstory

-1.06

OGND

-0.90

 Infórmanos

-0.88

sizeCache

-0.86

VYMaps

-0.81

MLLoader

-0.80

 Meksiku

-0.80

RTGC

-0.78

MigrationBuilder

-0.77

localctx

-0.77

POSITIVE LOGITS

 purpose

0.30

/*++

0.28

Abstract

0.28

laş

0.27

 sozinho

0.26

Достоинства

0.24

 works

0.24

 jang

0.24

 written

0.23

can

0.23

Activations Density 0.034%

references to pharmaceutical and medical terms

No Comments

No Known Activations