INDEX

Explanations

medical and health-related terms and concepts

oai_token-act-pair · gpt-3.5-turbo

terms and concepts related to medical and health topics

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

region

-0.64

*/(

-0.63

 peaks

-0.60

semble

-0.60

 contiguous

-0.59

 Horn

-0.59

lying

-0.59

standing

-0.58

Dot

-0.57

 Prosecut

-0.57

POSITIVE LOGITS

ysis

0.85

onics

0.73

alys

0.71

obia

0.70

ocaust

0.69

 ashore

0.69

aceutical

0.68

iologist

0.67

xon

0.67

ouses

0.67

Activations Density 0.341%

medical and health-related terms and concepts

terms and concepts related to medical and health topics

No Comments

No Known Activations