INDEX

Explanations

words related to diseases and health conditions

oai_token-act-pair · gpt-3.5-turbo

recurrent references to a specific disease

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

anian

-0.74

ruce

-0.71

ramid

-0.70

adra

-0.68

ilts

-0.67

KEN

-0.67

raltar

-0.66

agall

-0.66

gur

-0.66

assy

-0.65

POSITIVE LOGITS

 disease

1.06

 diseases

0.98

 outbreak

0.93

 Immun

0.87

 Disease

0.86

 outbreaks

0.86

 Diseases

0.85

 infections

0.83

 infection

0.82

 Outbreak

0.81

Activations Density 0.011%

words related to diseases and health conditions

recurrent references to a specific disease

No Comments

No Known Activations