INDEX

Explanations

parenthetical asides and interruptions.

oai_token-act-pair · gemini-2.0-flash

em dashes

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_7/width_16k/average_l0_70

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

--$

-0.48

oredCriteria

-0.43

DEPORTES

-0.43

--

-0.42

kranz

-0.42

--

-0.42

 ModelExpression

-0.42

ske

-0.42

protoc

-0.41

GetAsync

-0.41

POSITIVE LOGITS

 propOrder

0.77

DockStyle

0.74

AndEndTag

0.72

MLLoader

0.72

 WaitForSeconds

0.66

 AssemblyCompany

0.60

tisseur

0.60

 Meille

0.59

usermodel

0.58

doria

0.58

Activations Density 1.598%

parenthetical asides and interruptions.

em dashes

No Comments

No Known Activations