INDEX

Explanations

expressions that emphasize significant insights or takeaways

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

Follows "one," "this," or "the" and describes a concept

np_acts-logits-general · gemini-2.0-flash

identifying features or aspects

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-16k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

postmedia

-0.53

 officielles

-0.51

 laaj

-0.50

̀nh

-0.48

 suffices

-0.47

 peristiwa

-0.45

 anteriore

-0.44

 reicht

-0.44

 suffice

-0.44

花板

-0.43

POSITIVE LOGITS

 feature

0.83

 caractéristique

0.80

 aspect

0.79

 characteristic

0.78

 característica

0.77

feature

0.75

 FEATURE

0.74

queryInterface

0.72

这点

0.70

 Aspect

0.68

Activations Density 0.506%

expressions that emphasize significant insights or takeaways

Follows "one," "this," or "the" and describes a concept

identifying features or aspects

No Comments

No Known Activations