INDEX

Explanations

words related to product recalls and safety issues, such as reports, fires, injuries, and hazards

oai_token-act-pair · gpt-3.5-turbo

information regarding product recalls and safety notices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 passionately

-0.80

 persuasion

-0.78

 charisma

-0.78

iberal

-0.77

 ambition

-0.76

 wedd

-0.76

liber

-0.76

 poet

-0.76

 polygamy

-0.76

 democrat

-0.76

POSITIVE LOGITS

 defective

1.34

 faulty

1.24

ffect

1.21

 recall

1.18

 recalled

1.14

 manufacturer

1.07

ufact

1.06

 recalls

1.06

mone

1.04

 malfunction

1.01

Activations Density 0.517%

words related to product recalls and safety issues, such as reports, fires, injuries, and hazards

information regarding product recalls and safety notices

No Comments

No Known Activations