INDEX

Explanations

words related to books, authors, and literary 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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Bots

-0.70

ilitary

-0.67

00200000

-0.66

Yin

-0.66

 twitch

-0.65

hod

-0.63

xon

-0.63

asonic

-0.62

 Santiago

-0.59

ptions

-0.58

POSITIVE LOGITS

stores

1.43

seller

1.27

marks

1.17

shop

1.17

worm

1.12

worms

1.06

marked

1.05

sell

1.05

cases

0.99

 manuscript

0.98

Activations Density 0.633%

words related to books, authors, and literary activities

No Comments

No Known Activations