INDEX

Explanations

references to figures and their accompanying data in scientific documents

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

Letter followed by parenthesis and a hash

np_acts-logits-general · gemini-2.0-flash

citation markers

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

web

-0.38

큼

-0.36

océan

-0.36

 bater

-0.36

zy

-0.36

 Pring

-0.35

 αλ

-0.35

retirement

-0.33

 Defense

-0.33

POSITIVE LOGITS

GEBURTSDATUM

1.11

parsedMessage

0.89

ArgsConstructor

0.82

μβρίου

0.79

IsMutable

0.74

 strerror

0.74

NOPQRST

0.72

ViewFeatures

0.69

ertale

0.68

SharedCtor

0.68

Activations Density 0.558%

references to figures and their accompanying data in scientific documents

Letter followed by parenthesis and a hash

citation markers

No Comments

No Known Activations