INDEX

Explanations

phrases related to scams and deceptive practices

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

chwitz

-0.82

kind

-0.81

 largeDownload

-0.80

aldi

-0.79

issance

-0.68

anamo

-0.66

ghazi

-0.66

cember

-0.66

charged

-0.66

Eva

-0.65

POSITIVE LOGITS

 contemplate

0.95

 bother

0.94

 withstand

0.84

 blush

0.83

 survive

0.81

 compete

0.81

 sustain

0.81

 resist

0.80

 swallow

0.80

 waste

0.80

Activations Density 0.082%

phrases related to scams and deceptive practices

No Comments

No Known Activations