INDEX

Explanations

instances of the word "some" followed by different types of descriptions or nouns

oai_token-act-pair · gpt-3.5-turbo

the presence of the word "some" and variations indicating quantity or count

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

ukong

-0.67

tick

-0.65

 Username

-0.65

eston

-0.64

ession

-0.64

scrib

-0.64

vantage

-0.63

proc

-0.63

aeda

-0.63

APTER

-0.63

POSITIVE LOGITS

 downright

1.01

 lesser

0.87

 others

0.87

 exceptions

0.83

 importantly

0.81

 dozen

0.79

of

0.77

 thats

0.77

 else

0.77

 smaller

0.77

Activations Density 0.147%

instances of the word "some" followed by different types of descriptions or nouns

the presence of the word "some" and variations indicating quantity or count

No Comments

No Known Activations