INDEX

Explanations

occurrences of programming-related functions and error handling statements

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

Code, printing, or programming-related words

np_acts-logits-general · gemini-2.0-flash

code snippets and technical terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 (...)

-0.48

Hol

-0.46

 potentially

-0.46

?!

-0.46

potentially

-0.42

«

-0.42

 (…)

-0.42

пти

-0.40

ry

-0.40

Hu

-0.39

POSITIVE LOGITS

GEBURTSDATUM

1.09

 HasFactory

1.04

 defaultstate

1.00

########.

0.96

లాలు

0.96

riwal

0.88

}{@

0.87

complexContent

0.86

ArrowToggle

0.84

RTLD

0.83

Activations Density 0.057%

occurrences of programming-related functions and error handling statements

Code, printing, or programming-related words

code snippets and technical terms

No Comments

No Known Activations