INDEX

Explanations

programming specific terminology related to object-oriented features and method definitions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

keywords and code identifiers

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

Negative Logits

Diweddarwch

-0.69

RenderAtEndOf

-0.67

itattu

-0.64

tagHelperRunner

-0.63

DockStyle

-0.63

 himo

-0.60

󠁣

-0.59

󠁮

-0.59

⮕

-0.58

 الاطلاع

-0.58

POSITIVE LOGITS

 Miß

0.47

luß

0.47

 quelquefois

0.46

 skall

0.44

faßt

0.43

 idéia

0.43

RequestMethod

0.42

 mußten

0.42

 particolari

0.41

ResumeLayout

0.41

Activations Density 0.023%

programming specific terminology related to object-oriented features and method definitions

keywords and code identifiers

No Known Activations