INDEX

Explanations

mentions of new developments or innovations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

istine

-0.83

yip

-0.79

ocene

-0.78

OPA

-0.76

UMP

-0.74

amus

-0.73

aminer

-0.73

acebook

-0.73

xual

-0.73

osaurus

-0.71

POSITIVE LOGITS

bies

1.40

 arrivals

1.26

 entrants

1.20

 developments

1.12

 additions

1.11

bie

1.11

 discoveries

1.08

 generations

1.06

 hires

1.05

 recruits

1.01

Activations Density 0.069%

mentions of new developments or innovations

No Comments

No Known Activations