INDEX

Explanations

lead authors and researchers in scientific studies

oai_token-act-pair · gpt-3.5-turbo

mentions of lead authors in research studies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

URA

-0.72

pmwiki

-0.64

 domination

-0.63

 satur

-0.61

conservancy

-0.59

 Celt

-0.58

 fodder

-0.58

 Vengeance

-0.58

 cannon

-0.58

 blight

-0.57

POSITIVE LOGITS

hesis

1.00

essors

0.86

heses

0.86

authors

0.78

 authors

0.74

ilyn

0.73

 Professor

0.73

atively

0.72

 Investigator

0.72

 thereof

0.70

Activations Density 0.121%

lead authors and researchers in scientific studies

mentions of lead authors in research studies

No Comments

No Known Activations