INDEX

Explanations

phrases related to new technology or products

oai_token-act-pair · gpt-3.5-turbo

phrases introducing descriptions or explanations of various subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ONE

-0.66

"},"

-0.61

 Waiting

-0.61

course

-0.60

aq

-0.57

idious

-0.57

 Situation

-0.56

 Intermediate

-0.56

æĿ

-0.55

 COUR

-0.55

POSITIVE LOGITS

 utilizes

1.23

 incorporates

1.20

 combines

1.20

 debuted

1.18

 integrates

1.16

 collects

1.16

 originated

1.15

 specializes

1.14

 connects

1.13

 operates

1.12

Activations Density 0.211%

phrases related to new technology or products

phrases introducing descriptions or explanations of various subjects

No Comments

No Known Activations