INDEX

Explanations

instances of a person's awareness or knowledge about certain topics, events, or circumstances

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

uckles

-0.76

ibaba

-0.72

soever

-0.72

wark

-0.67

phabet

-0.67

ulo

-0.67

itte

-0.67

arters

-0.66

quer

-0.66

estyles

-0.66

POSITIVE LOGITS

 impending

1.39

 dangers

1.30

 existence

1.25

 danger

1.06

 plight

1.04

 pitfalls

1.03

 looming

1.03

 whereabouts

1.03

 possibility

1.03

 potential

1.00

Activations Density 0.207%

instances of a person's awareness or knowledge about certain topics, events, or circumstances

No Comments

No Known Activations