INDEX

Explanations

scientific terms and references in research publications

oai_token-act-pair · gpt-3.5-turbo

references to academic publishing and peer-reviewed research articles

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

 Patriot

-0.82

 charms

-0.79

 charm

-0.69

apon

-0.69

 Stronghold

-0.68

 totem

-0.67

ania

-0.67

jer

-0.66

 deserted

-0.66

tsky

-0.66

POSITIVE LOGITS

 citation

1.00

ournals

0.98

 PubMed

0.96

 journals

0.96

 biomedical

0.93

 literature

0.92

 publication

0.90

 papers

0.90

 Citation

0.89

 peer

0.89

Activations Density 0.162%

scientific terms and references in research publications

references to academic publishing and peer-reviewed research articles

No Comments

No Known Activations