INDEX

Explanations

terms related to quantitative changes and dependencies

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

formik

-0.65

 essentially

-0.61

SuppressMessage

-0.61

 zwar

-0.57

⏪

-0.56

 basically

-0.56

 clearly

-0.56

NameInMap

-0.55

 truly

-0.54

 Exacts

-0.53

POSITIVE LOGITS

StoryboardSegue

0.77

 للمعارف

0.64

 nguyễn

0.62

 MaterialApp

0.59

Portail

0.59

 quadrature

0.59

RectangleBorder

0.57

σμο

0.57

 GetEnumerator

0.57

Decomposition

0.57

Activations Density 0.983%

terms related to quantitative changes and dependencies

No Comments

No Known Activations