INDEX

Explanations

Obama-related 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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

Cth

-0.75

 Edinburgh

-0.73

TOR

-0.70

 ILCS

-0.68

asma

-0.64

essional

-0.63

Brave

-0.62

tml

-0.61

 Kuala

-0.61

upper

-0.61

POSITIVE LOGITS

Care

1.02

 appoint

1.01

omics

0.98

 administration

0.96

 Administration

0.92

 Jinping

0.89

 aides

0.86

 vetoed

0.86

 supporters

0.85

care

0.84

Activations Density 1.400%

Obama-related phrases

No Comments

No Known Activations