INDEX

Explanations

instances where the phrase "If you want to" is followed by an action or instruction

oai_token-act-pair · gpt-3.5-turbo

the phrase "if you want to" and its variations

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

-0.83

lies

-0.75

 Said

-0.70

quickShipAvailable

-0.68

Lag

-0.66

 Says

-0.65

typ

-0.65

rising

-0.65

ById

-0.65

Roy

-0.65

POSITIVE LOGITS

 maximize

1.11

 pursue

1.07

 emulate

1.07

 indulge

1.07

 partake

1.05

 participate

1.05

 avoid

1.05

 conserve

1.02

 broaden

1.01

 keep

1.00

Activations Density 0.079%

instances where the phrase "If you want to" is followed by an action or instruction

the phrase "if you want to" and its variations

No Comments

No Known Activations