INDEX

Explanations

publication information, specifically locations and dates, like "New York, 1997"

oai_token-act-pair · gemini-2.0-flash

gibberish after publication or incorporation information

np_token-act-pair-logits · gemini-2.0-flash

numbers

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_16/width_16k/average_l0_10

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

StoryboardSegue

-0.65

хьтан

-0.57

 accro

-0.56

 Waray

-0.54

OrBuilder

-0.53

 smokes

-0.52

 strdup

-0.52

 ModelExpression

-0.52

 pulumi

-0.50

ObjectContext

-0.49

POSITIVE LOGITS

enderror

0.52

didSet

0.51

MLLoader

0.51

packer

0.51

showMessage

0.51

Jegyzetek

0.50

ellees

0.50

boring

0.46

JsonInclude

0.46

clock

0.45

Activations Density 0.648%

publication information, specifically locations and dates, like "New York, 1997"

gibberish after publication or incorporation information

numbers

No Comments

No Known Activations