INDEX

Explanations

discourse markers that shift the topic or introduce a new point.

oai_token-act-pair · gemini-2.0-flash

gibberish tokens

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

discourse markers

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

ity

-0.53

SCAN

-0.46

 markets

-0.43

 SCAN

-0.42

Scan

-0.42

skip

-0.42

ecuzione

-0.42

澄

-0.42

はじめに

-0.42

ENE

-0.41

POSITIVE LOGITS

 dAtA

0.68

RegressionTest

0.66

OGND

0.64

 незавершена

0.63

 Roskov

0.61

GEBURTSDATUM

0.60

AndEndTag

0.59

IVEREF

0.57

keyColumn

0.57

ContentAsync

0.56

Activations Density 2.314%

discourse markers that shift the topic or introduce a new point.

gibberish tokens

discourse markers

No Comments

No Known Activations