INDEX

Explanations

references to prestigious entertainment awards and events

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ItemBackground

-0.77

 EnglishChoose

-0.66

TagHelper

-0.59

 समीक्षाओं

-0.59

 ostavi

-0.57

ształ

-0.55

 Мексичка

-0.55

lapsingToolbar

-0.55

nocache

-0.55

CodeAttribute

-0.55

POSITIVE LOGITS

 World

1.08

 Olympics

0.93

 Oscars

0.84

 Wimbledon

0.80

NBA

0.77

 Super

0.76

 Olympic

0.72

 Gramm

0.72

 National

0.72

 Worlds

0.70

Activations Density 0.269%

references to prestigious entertainment awards and events

No Comments

No Known Activations