INDEX

Explanations

words related to criminal activities and events

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

otle

-0.82

agna

-0.71

zona

-0.71

 Sparrow

-0.63

 Wooden

-0.63

agne

-0.62

 Heist

-0.61

 Defenders

-0.61

 Clarkson

-0.61

 Devi

-0.61

POSITIVE LOGITS

geon

1.24

geons

1.18

geoning

1.10

gent

1.05

iosity

1.01

ges

0.97

gew

0.96

rible

0.95

ging

0.95

rant

0.93

Activations Density 2.675%

words related to criminal activities and events

No Comments

No Known Activations