INDEX

Explanations

punctuation marks and programming syntax elements, indicating it is analyzing code

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

Code or special characters

np_acts-logits-general · gemini-2.0-flash

PDE to a point

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 nous

-0.52

avis

-0.50

 pelo

-0.48

sé

-0.47

der

-0.45

bek

-0.44

He

-0.44

({

-0.44

POSITIVE LOGITS

 myſelf

1.14

 AssemblyTitle

0.99

 Houſe

0.98

 Normdatei

0.95

 Monfieur

0.90

 themſelves

0.90

 himſelf

0.90

ſelf

0.88

 doubtnut

0.88

 pleaſure

0.87

Activations Density 0.120%

punctuation marks and programming syntax elements, indicating it is analyzing code

Code or special characters

PDE to a point

No Comments

No Known Activations