INDEX

Explanations

references to view-related elements or templates in a programming context

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

file paths and rendering

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

getAs

-0.35

ciptakan

-0.35

 float

-0.34

 Nied

-0.34

 Selbst

-0.34

 diser

-0.33

 Schra

-0.33

 Selb

-0.33

 once

-0.32

MemoryWarning

-0.32

POSITIVE LOGITS

ViewFeatures

0.61

IntoConstraints

0.55

DataBinding

0.54

 invokingState

0.54

IUrlHelper

0.52

 Италијани

0.51

Personensuche

0.51

CopyWith

0.50

ByVersion

0.50

 HttpNotFound

0.50

Activations Density 0.001%

references to view-related elements or templates in a programming context

file paths and rendering

No Comments

No Known Activations