INDEX

Explanations

words related to social class and descriptors of groups of people

oai_token-act-pair · gemini-2.0-flash

non-English tokens and technical terms

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ur

-0.42

 Wood

-0.41

 certe

-0.41

ensky

-0.41

 Tyler

-0.40

зок

-0.39

mat

-0.39

 Feind

-0.38

nao

-0.38

 certo

-0.38

POSITIVE LOGITS

 مشين

1.14

 CreateTagHelper

1.06

 defaultstate

1.05

abestanden

1.02

 nakalista

0.91

 تانيه

0.84

\{\\

0.82

 betweenstory

0.81

脚注の使い方

0.79

StructEnd

0.79

Activations Density 1.944%

words related to social class and descriptors of groups of people

non-English tokens and technical terms

No Comments

No Known Activations

words related to social class and descriptors of groups of people

non-English tokens and technical terms

No Comments

No Known Activations