INDEX

Explanations

names of people or organizations

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

[&

-0.80

è¦ļéĨĴ

-0.74

 guiActiveUnfocused

-0.66

 Skydragon

-0.66

 Masquerade

-0.65

 Sigma

-0.64

 Cthulhu

-0.64

 thereof

-0.63

 Spac

-0.63

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.60

POSITIVE LOGITS

odore

1.18

jamin

1.13

anmar

1.07

anie

1.04

alyst

1.03

resa

1.02

vertising

1.02

ropolitan

1.01

withstanding

0.99

dinand

0.99

Activations Density 1.586%

names of people or organizations

No Comments

No Known Activations