INDEX

Explanations

phrases related to medical diagnoses

oai_token-act-pair · gpt-3.5-turbo

references to medical diagnoses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

sol

-0.71

 modesty

-0.70

perty

-0.68

atility

-0.67

perm

-0.67

de

-0.63

assi

-0.62

adish

-0.61

hire

-0.60

da

-0.60

POSITIVE LOGITS

 diagnosed

1.02

ostics

0.95

 Diagn

0.93

 diagnoses

0.90

 diagnosis

0.84

 diagn

0.77

 Symptoms

0.76

osis

0.76

ostic

0.74

omas

0.74

Activations Density 0.013%

phrases related to medical diagnoses

references to medical diagnoses

No Comments

No Known Activations