INDEX

Explanations

keywords related to titles or named entities

oai_token-act-pair · gpt-3.5-turbo

specific titles or names related to games, projects, or notable events

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

istically

-0.77

ded

-0.76

aneously

-0.74

 exception

-0.70

ctor

-0.70

ties

-0.70

onyms

-0.69

lier

-0.68

liest

-0.65

ished

-0.64

POSITIVE LOGITS

mith

1.53

peed

1.41

hift

1.37

pace

1.25

ystem

1.23

ilver

1.21

pread

1.18

afety

1.18

aurus

1.15

chool

1.15

Activations Density 0.235%

keywords related to titles or named entities

specific titles or names related to games, projects, or notable events

No Comments

No Known Activations