INDEX

Explanations

proper nouns related to the name "Lex."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "Lex" and variations of "flex."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

Â¢

-0.73

 parties

-0.65

 acknowled

-0.64

 Reincarn

-0.62

 untold

-0.62

 Parties

-0.62

OTS

-0.60

 payable

-0.60

orld

-0.59

 Rabbit

-0.59

POSITIVE LOGITS

ibly

1.36

ibility

1.28

ible

1.21

ibl

1.16

icon

1.16

ibilities

1.08

itime

1.05

ical

0.98

ician

0.96

erion

0.95

Activations Density 0.060%

proper nouns related to the name "Lex."

occurrences of the term "Lex" and variations of "flex."

No Comments

No Known Activations