INDEX

Explanations

words related to controversial or negative situations

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

backer

-0.72

ivities

-0.71

meal

-0.67

 Prosper

-0.67

 nomine

-0.66

bilt

-0.65

 virtue

-0.65

 Ceres

-0.64

 Perse

-0.63

 Galactic

-0.62

POSITIVE LOGITS

bing

1.36

yrinth

1.21

oard

1.18

ecause

1.17

elled

1.16

bles

1.15

bers

1.14

bed

1.14

ruary

1.13

dullah

1.11

Activations Density 1.775%

words related to controversial or negative situations

No Comments

No Known Activations