INDEX

Explanations

phrases related to dangerous or harmful situations

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

theless

-0.81

 Flavoring

-0.74

works

-0.72

 advoc

-0.67

work

-0.66

amac

-0.63

backs

-0.62

 prompts

-0.61

 gallery

-0.60

Ø¹

-0.58

POSITIVE LOGITS

bered

1.18

oths

1.10

far

0.99

iled

0.95

othes

0.93

ociated

0.87

apy

0.85

 much

0.85

oooo

0.85

ooo

0.83

Activations Density 0.069%

phrases related to dangerous or harmful situations

No Comments

No Known Activations