INDEX

Explanations

names of specific individuals

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

IPM

-0.77

 à¨

-0.74

 behavi

-0.73

 mathemat

-0.70

-0.68

 livest

-0.68

CLASSIFIED

-0.64

 hemor

-0.64

 gravitational

-0.63

assetsadobe

-0.63

POSITIVE LOGITS

zon

0.99

kowski

0.94

rod

0.92

igan

0.88

enson

0.87

jan

0.85

bra

0.85

row

0.85

imoto

0.85

son

0.85

Activations Density 0.136%

names of specific individuals

No Comments

No Known Activations