INDEX

Explanations

programming-related terms and operations

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

gtk tree append

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

!*\

-0.65

AddTagHelper

-0.59

 يتيمه

-0.58

RenderAtEndOf

-0.56

EndContext

-0.55

MessageTagHelper

-0.55

 disambiguazione

-0.55

abestanden

-0.55

RunWith

-0.54

 kasarigan

-0.54

POSITIVE LOGITS

WidgetItem

0.45

 بلو

0.35

jerc

0.33

 yearly

0.33

 voire

0.33

artment

0.33

риста

0.33

hnia

0.33

 insertion

0.33

PARTMENT

0.32

Activations Density 0.029%

programming-related terms and operations

gtk tree append

No Comments

No Known Activations