INDEX

Explanations

phrases that refer to ongoing trends or actions happening in a sequence

oai_token-act-pair · gpt-3.5-turbo

phrases indicating ongoing trends or developments

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.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

 Pict

-0.69

 Gallery

-0.67

Pak

-0.64

Gy

-0.61

 Doctrine

-0.60

 northeastern

-0.59

river

-0.59

 Speaking

-0.58

Thy

-0.56

aq

-0.56

POSITIVE LOGITS

 consumes

0.89

 extends

0.87

 awaits

0.87

soever

0.86

 arose

0.86

 occurs

0.85

 encompasses

0.84

 haun

0.84

 happens

0.84

 undermines

0.84

Activations Density 0.126%

phrases that refer to ongoing trends or actions happening in a sequence

phrases indicating ongoing trends or developments

No Comments

No Known Activations