INDEX

Explanations

HTML/CSS class and style attributes

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

HTML class attributes and mathematical subscripts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

 незавершена

-0.91

Efq

-0.88

 ſind

-0.83

 doubtnut

-0.82

 Anſ

-0.82

 pleaſure

-0.80

 Shakspeare

-0.80

 iſt

-0.78

 Majefty

-0.78

 Monfieur

-0.78

POSITIVE LOGITS

co

0.60

ch

0.54

fe

0.52

te

0.52

 dominar

0.49

 background

0.48

 Background

0.47

pre

0.46

 class

0.46

Pre

0.46

Activations Density 0.012%

HTML/CSS class and style attributes

HTML class attributes and mathematical subscripts

No Comments

No Known Activations