INDEX

Explanations

mentions of foreign entities or activities

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

illation

-0.89

hots

-0.82

byss

-0.81

adobe

-0.79

earable

-0.79

illy

-0.75

++++

-0.75

YR

-0.74

 DRAG

-0.74

 GOODMAN

-0.73

POSITIVE LOGITS

 nationals

1.20

 affairs

1.06

 countries

0.96

 dign

0.93

 nations

0.93

 shores

0.92

 invaders

0.92

 embassies

0.89

 languages

0.88

 governments

0.85

Activations Density 0.020%

mentions of foreign entities or activities

No Comments

No Known Activations