INDEX

Explanations

terms related to boxed sets or packages, especially in a context involving math or programming

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

boxed set or mathematical notation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

ThroughAttribute

-0.48

 مرئيه

-0.46

thello

-0.45

UrlHelper

-0.42

 mitsubishi

-0.41

 garcía

-0.41

 pelicula

-0.41

 mètre

-0.41

 charité

-0.40

éroport

-0.40

POSITIVE LOGITS

boxed

3.23

 boxed

1.48

fbox

1.02

box

0.91

Box

0.82

boxes

0.78

BOX

0.77

boxing

0.76

Box

0.76

BOX

0.73

Activations Density 0.010%

terms related to boxed sets or packages, especially in a context involving math or programming

boxed set or mathematical notation

No Comments

No Known Activations