INDEX

Explanations

code-related tokens

oai_token-act-pair · gemini-2.0-flash

programming syntax or coding terms

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

Code or UI layouts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tient

-0.52

 testified

-0.49

 testify

-0.47

PROGRAM

-0.46

 Goodman

-0.46

 Pollard

-0.45

opardy

-0.45

 Five

-0.45

ada

-0.44

pag

-0.43

POSITIVE LOGITS

+#+#

1.03

 Wikimedijinoj

0.98

 kasarigan

0.86

DeleteBehavior

0.84

ContentAsync

0.77

存于互联网档案馆

0.77

twimg

0.74

principalColumn

0.74

abestanden

0.73

ineno

0.73

Activations Density 0.141%

code-related tokens

programming syntax or coding terms

Code or UI layouts

No Comments

No Known Activations