INDEX

Explanations

medical treatments or interventions

oai_token-act-pair · gpt-3.5-turbo

references to treatment in a medical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

bour

-0.67

Kyr

-0.66

 compr

-0.66

 Eater

-0.64

fry

-0.64

ross

-0.63

 juicy

-0.63

iak

-0.63

erd

-0.62

 Geological

-0.62

POSITIVE LOGITS

 regimen

1.19

 therapies

1.03

 treatments

0.98

 protocols

0.94

 treatment

0.92

 therapy

0.90

 options

0.89

ilitating

0.89

 effectiveness

0.83

utic

0.83

Activations Density 0.041%

medical treatments or interventions

references to treatment in a medical context

No Comments

No Known Activations