INDEX

Explanations

names of people, places, and organizations related to news events

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

 Debor

-0.81

âĶģ

-0.72

 Syndicate

-0.71

orpor

-0.70

 Triumph

-0.68

é¾įå¥ĳå£«

-0.65

 Customs

-0.64

 Tonight

-0.64

 Revelations

-0.63

å§«

-0.63

POSITIVE LOGITS

kees

1.01

eter

0.94

quet

0.90

ilee

0.90

vel

0.86

isin

0.84

hao

0.83

jo

0.82

issance

0.80

adena

0.80

Activations Density 0.015%

names of people, places, and organizations related to news events

No Comments

No Known Activations