INDEX

Explanations

programming syntax and structure, particularly in C/C++ and similar languages

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

programming and technical terms

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

httphttps

-0.62

AndEndTag

-0.60

 snippetHide

-0.53

findpost

-0.51

TagMode

-0.49

geslacht

-0.48

numerusform

-0.47

WebElementEntity

-0.47

webElementXpaths

-0.46

apimachinery

-0.46

POSITIVE LOGITS

etc

0.49

 demografica

0.41

などなど

0.41

Dichloropropene

0.36

ecc

0.36

 SEDS

0.36

等等

0.35

 perche

0.34

ECI

0.33

CELLANEOUS

0.33

Activations Density 0.017%

programming syntax and structure, particularly in C/C++ and similar languages

programming and technical terms

No Comments

No Known Activations