INDEX

Explanations

phrases related to pushing boundaries and challenging authority

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

quel

-0.81

SourceFile

-0.77

reau

-0.71

ily

-0.71

HK

-0.70

hester

-0.70

ãĥĹ

-0.69

>>>>>>>>

-0.69

rang

-0.68

eus

-0.67

POSITIVE LOGITS

 civilisation

0.87

 legality

0.87

 civilization

0.84

 decency

0.81

 sanity

0.81

 darkness

0.80

 obscurity

0.79

 academia

0.78

 humanity

0.76

 adulthood

0.73

Activations Density 0.181%

phrases related to pushing boundaries and challenging authority

No Comments

No Known Activations