INDEX

Explanations

syntax and structure related to code or programming elements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

packed enum fields

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ChildScrollView

-0.70

⎦

-0.49

ArgsConstructor

-0.44

QMetaType

-0.43

BindingResult

-0.42

 OMITBAD

-0.42

 onPostExecute

-0.41

PyExc

-0.40

'][]

-0.40

complexContent

-0.39

POSITIVE LOGITS

ACHUSET

0.43

ọa

0.41

ITURE

0.40

 maso

0.40

Rot

0.38

 ब्रेकडाउन

0.38

 Chuan

0.38

EndGlobalSection

0.38

 TType

0.37

+#+#

0.36

Activations Density 0.003%

syntax and structure related to code or programming elements

packed enum fields

No Comments

No Known Activations