INDEX

Explanations

technology-related terms and keywords related to digital security and vulnerabilities

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

staking

-0.58

dragon

-0.57

 Barron

-0.57

DragonMagazine

-0.56

eks

-0.56

[&

-0.55

 secretion

-0.54

 branches

-0.54

 frequencies

-0.53

 Advice

-0.53

POSITIVE LOGITS

Ħ¢

0.92

 manageable

0.70

 cheaply

0.69

handed

0.68

readable

0.66

accessible

0.65

liv

0.64

 easier

0.62

="#

0.62

aredevil

0.62

Activations Density 1.723%

technology-related terms and keywords related to digital security and vulnerabilities

No Comments

No Known Activations