INDEX

Explanations

code-related terms, particularly those related to image processing and data structures

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

width, height, dimensions, size

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

 positive

-0.36

positive

-0.33

évaluateur

-0.32

 push

-0.32

Positive

-0.31

 Positive

-0.31

AnchorTagHelper

-0.29

 ostavi

-0.29

 positiven

-0.28

 Sven

-0.28

POSITIVE LOGITS

 utafitiHapana

0.63

 esternos

0.60

الحياه

0.58

farwyddwr

0.57

 CanadaChoose

0.55

 ddelweddau

0.54

 protoimpl

0.54

 MainAxisSize

0.54

 getHeight

0.53

 ddelwed

0.53

Activations Density 0.017%

code-related terms, particularly those related to image processing and data structures

width, height, dimensions, size

No Comments

No Known Activations