INDEX

Explanations

structural elements of an HTML document

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

or characters following bracketed numbers

np_acts-logits-general · gemini-2.0-flash

numbers followed by units or codes

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

ing

-0.50

filme

-0.47

oti

-0.45

 volontà

-0.44

 calldata

-0.41

voorbeeld

-0.41

kiksi

-0.40

üyü

-0.40

 murni

-0.39

 intestino

-0.39

POSITIVE LOGITS

GraphicsUnit

0.83

<h1>

0.81

itespace

0.74

LEncoder

0.68

abestanden

0.65

rógeno

0.65

ThroughAttribute

0.63

AddTagHelper

0.62

#+#

0.62

 onCreateView

0.61

Activations Density 0.044%

structural elements of an HTML document

or characters following bracketed numbers

numbers followed by units or codes

No Comments

No Known Activations