INDEX

Explanations

names of objects or concepts that are plural

oai_token-act-pair · gpt-3.5-turbo

nouns or terms associated with various categories or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

rawdownloadcloneembedreportprint

-0.68

00200000

-0.65

 Tuls

-0.64

bal

-0.63

ABLE

-0.63

Scient

-0.62

Tur

-0.62

Wah

-0.61

ãĥ£

-0.60

lihood

-0.60

POSITIVE LOGITS

ettings

1.20

chool

1.07

etting

1.04

hip

1.03

igmatic

1.01

ilver

0.99

mith

0.99

avers

0.98

paces

0.97

cale

0.96

Activations Density 0.642%

names of objects or concepts that are plural

nouns or terms associated with various categories or classifications

No Comments

No Known Activations