INDEX

Explanations

words that relate to libraries, research, and the news

oai_token-act-pair · gemini-2.0-flash

library

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Efq

-0.90

 Anſ

-0.88

 auffi

-0.84

ſelves

-0.83

 ―――――

-0.82

 Majefty

-0.82

 itſelf

-0.82

 myſelf

-0.81

 Jefus

-0.80

ſelf

-0.80

POSITIVE LOGITS

de

0.42

ar

0.42

de

0.41

อด

0.39

wet

0.39

 való

0.38

ķ

0.38

引

0.37

 materiál

0.36

REFERENCE

0.35

Activations Density 1.913%

words that relate to libraries, research, and the news

library

No Comments

No Known Activations