INDEX

Explanations

terminology related to prototyping and stereotypes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

 attendant

-0.69

 attendants

-0.66

 Cheong

-0.63

FO

-0.62

 physician

-0.60

 rush

-0.59

IRO

-0.59

 Frey

-0.59

Rae

-0.58

 GENERAL

-0.57

POSITIVE LOGITS

ical

1.38

otyp

1.18

ipl

1.16

ically

1.09

ing

1.08

ed

1.05

ĭ

1.02

ifact

1.00

itive

0.99

inary

0.98

Activations Density 0.003%

terminology related to prototyping and stereotypes

No Comments

No Known Activations