INDEX

Explanations

references to work, employment, or professional activities

oai_token-act-pair · gpt-4o-mini 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

emonic

-0.80

 constitu

-0.73

xual

-0.67

ĪĴ

-0.66

 Bubble

-0.66

angular

-0.65

ople

-0.64

 Augustus

-0.64

anus

-0.64

iren

-0.61

POSITIVE LOGITS

 tirelessly

1.10

hops

1.01

 ethic

1.01

 diligently

1.00

 overtime

0.94

 collabor

0.94

bench

0.91

aday

0.84

paces

0.83

 undercover

0.82

Activations Density 0.063%

references to work, employment, or professional activities

No Comments

No Known Activations