INDEX

Explanations

code snippets, figures, and tables.

oai_token-act-pair · gemini-2.0-flash

figure and table references

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

 Loren

-0.66

 Affleck

-0.65

······

-0.65

mographie

-0.63

ISupport

-0.63

 Batista

-0.63

Biographie

-0.60

SourceChecksum

-0.59

cote

-0.58

endorong

-0.58

POSITIVE LOGITS

 Eddy

0.60

 Norr

0.59

 Elton

0.57

 McQueen

0.57

Eddy

0.57

 Eccl

0.57

Gok

0.57

 Nikki

0.56

 Jamestown

0.56

Daw

0.55

Activations Density 5.619%

code snippets, figures, and tables.

figure and table references

No Comments

No Known Activations