INDEX

Explanations

adjectives and words indicating ownership or origin

oai_token-act-pair · gemini-2.0-flash

tokens related to important entities or titles

np_token-act-pair-logits · gpt-4o-mini

Foreign words

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ſeveral

-0.91

Efq

-0.90

 itſelf

-0.88

 reaſon

-0.87

 Monfieur

-0.85

 myſelf

-0.85

 houſe

-0.84

 endregion

-0.82

 raiſ

-0.80

 Anſ

-0.80

POSITIVE LOGITS

出版年

0.49

 Erişim

0.46

Chham

0.44

гата

0.44

Архівовано

0.43

 nombreux

0.43

 also

0.43

Dtos

0.42

참고

0.42

 متعلقه

0.41

Activations Density 0.466%

adjectives and words indicating ownership or origin

tokens related to important entities or titles

Foreign words

No Comments

No Known Activations