INDEX

Explanations

code-related keywords and structures

oai_token-act-pair · gemini-2.0-flash

programming keywords and structure

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

public

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

ct

-0.57

ta

-0.56

om

-0.55

al

-0.55

pt

-0.54

da

-0.54

ga

-0.53

ro

-0.53

ag

-0.52

so

-0.52

POSITIVE LOGITS

 كومونز

0.72

araman

0.69

esterday

0.69

 متعلقه

0.68

ientôt

0.68

FTFY

0.67

RenderAtEndOf

0.66

prehensive

0.66

etheless

0.66

roduction

0.66

Activations Density 1.618%

code-related keywords and structures

programming keywords and structure

public

No Comments

No Known Activations