INDEX

Explanations

fields and properties related to programming requests and APIs

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

public declarations

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

 queſta

-1.08

iſchen

-1.07

Vidite

-1.05



-1.03

ſſung

-1.03

niſſe

-1.02

 geſch

-1.01

ſicht

-1.01

ſchaft

-1.00

 erſt

-1.00

POSITIVE LOGITS

0.48

0.45

0.43

0.39

public

0.38

The

0.38

But

0.38

if

0.38

 function

0.36

Activations Density 0.007%

fields and properties related to programming requests and APIs

public declarations

No Comments

No Known Activations