INDEX

Explanations

proper nouns

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

GST

-0.75

 Corpus

-0.68

 Clyde

-0.65

 eBay

-0.65

Uni

-0.65

 Territ

-0.65

vic

-0.65

 Patreon

-0.65

 Costco

-0.64

NXT

-0.64

POSITIVE LOGITS

acci

1.10

irez

0.94

enegger

0.93

olini

0.90

uga

0.90

enson

0.86

ison

0.84

anyahu

0.83

ega

0.80

inson

0.80

Activations Density 0.193%

proper nouns

No Comments

No Known Activations