INDEX

Explanations

concepts related to underlying principles or mechanisms within various contexts

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

"Underlying" or "fundamental" concepts

np_acts-logits-general · gemini-2.0-flash

essential nature

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Lightboxes

-0.62

 anzi

-0.60

 passée

-0.58

 NSCoder

-0.57

zola

-0.57

 Commencez

-0.56

antMatchers

-0.56

 ocas

-0.56

 vean

-0.55

MIDDLEWARE

-0.55

POSITIVE LOGITS

 fundamental

0.84

 underlying

0.82

 Underlying

0.81

 core

0.81

 fundamentals

0.81

Underlying

0.79

 grund

0.78

 basic

0.78

 fondamental

0.76

underlying

0.76

Activations Density 0.254%

concepts related to underlying principles or mechanisms within various contexts

"Underlying" or "fundamental" concepts

essential nature

No Comments

No Known Activations