INDEX

Explanations

references to news reporting and media agency mentions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

mú

-0.47

 publicação

-0.46

OpenHelper

-0.46

sunt

-0.45

printLine

-0.43

 publication

-0.43

 blog

-0.42

анта

-0.42

 MotionEvent

-0.42

Revista

-0.42

POSITIVE LOGITS

 agency

1.05

 agencies

0.97

 Agency

0.96

 Agencies

0.95

 Associated

0.95

 agences

0.93

AFP

0.91

 wire

0.90

 Wires

0.90

 wires

0.89

Activations Density 0.123%

references to news reporting and media agency mentions

No Comments

No Known Activations