INDEX

Explanations

words related to abduction

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 virtue

-0.82

 nomine

-0.71

 Perse

-0.68

 Patriot

-0.62

 antidote

-0.61

 Ceres

-0.61

 Coco

-0.61

Magikarp

-0.61

 Sandy

-0.60

 Izan

-0.59

POSITIVE LOGITS

stract

1.32

yrinth

1.15

oard

1.09

bing

1.08

raham

1.06

ylon

1.04

dullah

1.02

urger

1.01

riel

1.01

ble

0.98

Activations Density 0.051%

words related to abduction

No Comments

No Known Activations