INDEX

Explanations

words related to scientific publications

oai_token-act-pair · gpt-3.5-turbo

acronyms and terms related to scientific research publications and journals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

recomm

-0.70

REDACTED

-0.69

 Recomm

-0.68

CFR

-0.67

ONSORED

-0.65

è¦ļéĨĴ

-0.65

displayText

-0.63

 notch

-0.62

 Ancients

-0.62

 scorp

-0.59

POSITIVE LOGITS

etooth

0.82

encia

0.75

emon

0.73

chel

0.73

nce

0.73

hett

0.72

eland

0.72

idi

0.71

ocene

0.70

ppe

0.70

Activations Density 0.177%

words related to scientific publications

acronyms and terms related to scientific research publications and journals

No Comments

No Known Activations