INDEX

Explanations

references to different forms of architecture

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

SourceChecksum

-0.84

 CURIAM

-0.74

 >=",

-0.73

DotNetBar

-0.69

 كومونز

-0.67

ExtendWith

-0.67

 виправивши

-0.67

fromnode

-0.66

awtextra

-0.66

 glyphicon

-0.66

POSITIVE LOGITS

 compatibles

0.66

 incompatible

0.63

Slot

0.59

shape

0.57

 fidèle

0.57

 compatible

0.56

shaped

0.55

Architecture

0.55

Incompatible

0.55

IBM

0.54

Activations Density 0.117%

references to different forms of architecture

No Comments

No Known Activations