INDEX

Explanations

Twitter-related content

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

cised

-0.69

 Scandinavian

-0.66

 refriger

-0.63

ricular

-0.62

 circumcision

-0.61

vantage

-0.61

cision

-0.60

minded

-0.60

ortium

-0.60

 Starr

-0.60

POSITIVE LOGITS

storms

0.95

 hashtag

0.94

(@

0.92

@@@@@@@@

0.89

 hasht

0.88

"@

0.87

 feeds

0.85

username

0.84

Tweet

0.84

 Username

0.83

Activations Density 0.367%

Twitter-related content

No Comments

No Known Activations