INDEX

Explanations

references to books and reading-related 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_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

ilitary

-0.70

 Bots

-0.70

xon

-0.65

asonic

-0.64

00200000

-0.63

uppet

-0.61

hod

-0.60

Yin

-0.60

 twitch

-0.60

nel

-0.58

POSITIVE LOGITS

stores

1.35

seller

1.21

shop

1.13

 manuscript

1.12

marks

1.01

 books

1.00

 publisher

0.99

worm

0.98

 publishers

0.98

books

0.97

Activations Density 2.003%

references to books and reading-related activities

No Comments

No Known Activations