INDEX

Explanations

testing-related keywords and structure in code

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

Code testing contexts ("it", "test")

np_acts-logits-general · gemini-2.0-flash

test descriptions and code

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 المعيارى

-0.97

MigrationBuilder

-0.91

fjspx

-0.82

enumi

-0.81

 فريبيس

-0.81

 esternos

-0.81

contentLoaded

-0.79

IsMutable

-0.79

клопе

-0.78

تقاوى

-0.77

POSITIVE LOGITS

 “…

0.53

 disambiguazione

0.47

junit

0.46

xam

0.46

της

0.45

 should

0.45

 that

0.45

 could

0.44

reus

0.43

Suara

0.42

Activations Density 0.051%

testing-related keywords and structure in code

Code testing contexts ("it", "test")

test descriptions and code

No Comments

No Known Activations