INDEX

Explanations

words or phrases that are descriptive, past tense, and somewhat academic

oai_token-act-pair · gemini-2.0-flash

result

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

 AssemblyCulture

-1.04

MigrationBuilder

-1.01

expandindo

-1.00

ValueStyle

-0.99

parsedMessage

-0.98

 <=",

-0.98

 otomatig

-0.94

CloseOperation

-0.94

 للمعارف

-0.93

 CreateTagHelper

-0.91

POSITIVE LOGITS

in

0.56

to

0.55

0.54

two

0.50

by

0.50

end

0.48

立ち

0.47

 late

0.46

got

0.45

Activations Density 2.465%

words or phrases that are descriptive, past tense, and somewhat academic

result

No Comments

No Known Activations