INDEX

Explanations

medical and surgical terminologies

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

eworld

-0.80

erent

-0.75

Torrent

-0.75

rising

-0.71

bitious

-0.69

imeo

-0.65

reprene

-0.65

cakes

-0.65

=]

-0.64

liners

-0.63

POSITIVE LOGITS

 examination

1.24

 examinations

1.21

 technician

1.10

 scan

1.09

 scans

1.07

 Technician

1.03

 exam

1.00

 procedure

1.00

 scanners

0.98

 procedures

0.98

Activations Density 1.151%

medical and surgical terminologies

No Comments

No Known Activations