INDEX

Explanations

the word "have"

oai_token-act-pair · gpt-3.5-turbo

uses of the verb "have."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Apart

-0.73

catentry

-0.64

oshi

-0.59

icking

-0.58

 Crown

-0.55

eem

-0.55

arter

-0.54

TG

-0.54

 recalls

-0.54

 dodge

-0.53

POSITIVE LOGITS

 been

1.37

been

1.19

 Been

1.02

 gotten

1.00

 seen

0.93

 done

0.90

 undergone

0.89

 begun

0.88

 become

0.87

 gone

0.86

Activations Density 0.321%

the word "have"

uses of the verb "have."

No Comments

No Known Activations