INDEX

Explanations

security and technology-related words and phrases

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

 prints

-0.59

 fingerprints

-0.58

 substitutes

-0.56

 converter

-0.56

 antit

-0.55

 comple

-0.55

 Albania

-0.55

 Permanent

-0.55

Aph

-0.55

 blinded

-0.55

POSITIVE LOGITS

WASHINGTON

0.88

DEN

0.82

Gov

0.79

ccording

0.78

everal

0.78

BBC

0.77

ÂłÂłÂł

0.76

pmwiki

0.76

LOS

0.75

ente

0.74

Activations Density 0.138%

security and technology-related words and phrases

No Comments

No Known Activations