INDEX

Explanations

technical jargon and terms related to user interface design and programming

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

refresh and retrieval actions

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

awtextra

-0.76

GraphicsUnit

-0.71

ScopeManager

-0.66

rungsseite

-0.63

SharedCtor

-0.62

endphp

-0.61

tagHelperRunner

-0.60

 joaat

-0.57

AccessorTable

-0.56

ControllerAdvice

-0.55

POSITIVE LOGITS

ToRefresh

0.81

RefreshLayout

0.66

 OMITBAD

0.42

Refresh

0.38

Refreshing

0.33

 исход

0.33

Uwagi

0.33

 vlasy

0.32

require

0.31

 temporaire

0.30

Activations Density 0.002%

technical jargon and terms related to user interface design and programming

refresh and retrieval actions

No Comments

No Known Activations