INDEX

Explanations

regular expression syntax

oai_token-act-pair · gemini-2.0-flash

non-English words or strings

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

code and urls

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 WaitForSeconds

-0.56

ypeł

-0.48

 ongles

-0.46

 TextInputType

-0.44

 grå

-0.43

timewa

-0.43

 tradition

-0.42

engagent

-0.42

 Dyck

-0.41

 syndical

-0.41

POSITIVE LOGITS

uxxxx

0.76

CloseOperation

0.75

Personendaten

0.73

허

0.73

Personensuche

0.70

évaluateur

0.70

OGND

0.65

 AssemblyProduct

0.64

UrlResolution

0.63

 nahilalakip

0.63

Activations Density 2.395%

regular expression syntax

non-English words or strings

code and urls

No Comments

No Known Activations