INDEX

Explanations

scientific and medical terms related to health and research

oai_token-act-pair · gpt-3.5-turbo

references to medical treatments and their effects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

soDeliveryDate

-0.90

 acron

-0.75

 iceberg

-0.73

assetsadobe

-0.68

kos

-0.66

 sails

-0.65

 looms

-0.65

craft

-0.64

 plumbing

-0.64

 brim

-0.63

POSITIVE LOGITS

RESULTS

1.13

 (âĪĴ

1.12

 placebo

1.09

 decreased

1.07

 assessed

0.97

 statistically

0.95

 compared

0.93

ipop

0.93

 observed

0.91

 biomark

0.91

Activations Density 0.923%

scientific and medical terms related to health and research

references to medical treatments and their effects

No Comments

No Known Activations