INDEX

Explanations

phrases related to the inability to do something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "cannot."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

soDeliveryDate

-0.71

heavy

-0.66

bats

-0.64

senal

-0.64

ItemImage

-0.60

 Signs

-0.60

Tow

-0.60

 Born

-0.59

grounds

-0.59

Lif

-0.59

POSITIVE LOGITS

 afford

1.44

 comprehend

1.06

 feas

1.03

 tolerate

1.02

 reproduce

0.98

 penetrate

0.97

 necessarily

0.96

 adequately

0.95

 conceive

0.93

 withstand

0.93

Activations Density 0.047%

phrases related to the inability to do something

instances of the word "cannot."

No Comments

No Known Activations