INDEX

Explanations

names of news programs and networks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

Diweddarwch

-0.65

AddTagHelper

-0.52

webElementXpaths

-0.51

 Administrativna

-0.51

 JpaRepository

-0.49

StandardCharsets

-0.49

 SQLAlchemy

-0.48

 للمعارف

-0.47

protoimpl

-0.47

 nahilalakip

-0.46

POSITIVE LOGITS

TV

1.09

 television

1.06

tv

1.02

 televised

0.96

TV

0.91

 televis

0.89

 Television

0.87

television

0.84

 televisi

0.83

 broadcast

0.79

Activations Density 0.304%

names of news programs and networks

No Comments

No Known Activations