INDEX

Explanations

technical jargon related to apparatus and mechanisms

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

Formatting or dimension specifications

np_acts-logits-general · gemini-2.0-flash

dimensions and descriptive features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

+:+

-0.50

yet

-0.48

 despite

-0.48

 nevertheless

-0.43

 genuinely

-0.42

帰

-0.42

 formally

-0.42

 vetores

-0.42

who

-0.41

 whom

-0.41

POSITIVE LOGITS

 Adjustable

0.77

 Removable

0.75

 Mounted

0.72

 Stainless

0.70

 removable

0.70

 Optional

0.70

 Dimensions

0.69

 Mounting

0.66

 optional

0.66

 Diameter

0.66

Activations Density 0.435%

technical jargon related to apparatus and mechanisms

Formatting or dimension specifications

dimensions and descriptive features

No Comments

No Known Activations