INDEX

Explanations

terms related to clusters and specific codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 StatefulWidget

-0.75

numerusform

-0.66

Hozzáférés

-0.64

الحياه

-0.64

BeginContext

-0.62

WriteLiteral

-0.59

Extinguishing

-0.59

出版年

-0.56

 BoxFit

-0.56

weit

-0.56

POSITIVE LOGITS

 mentors

0.75

 mentor

0.69

IContainer

0.62

AppRoutingModule

0.62

oredCriteria

0.61

Geplaatst

0.61

 Roskov

0.60

 éto

0.60

Numerology

0.54

 himo

0.54

Activations Density 0.160%

terms related to clusters and specific codes

No Comments

No Known Activations