INDEX

Explanations

references to websites and sources of additional information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

tagHelperRunner

-0.92

 myſelf

-0.86

 Perſ

-0.83

 Monfieur

-0.83

 itſelf

-0.83

 שוליים

-0.80

PMailer

-0.79

 Anſ

-0.79

 poffible

-0.79

 Houſe

-0.79

POSITIVE LOGITS

www

0.72

 http

0.72

 https

0.62

web

0.54

<bos>

0.54

 online

0.54

http

0.53

www

0.52

 HERE

0.50

Activations Density 0.168%

references to websites and sources of additional information

No Comments

No Known Activations