INDEX

Explanations

phrases related to medical professionals, particularly doctors

oai_token-act-pair · gpt-3.5-turbo

instances of the word "doctors" and related medical terminology

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

 Territories

-0.90

 Flavoring

-0.74

Characters

-0.65

uble

-0.64

regate

-0.63

CHAT

-0.63

Campaign

-0.62

âĸ¬

-0.61

Topics

-0.61

TRUMP

-0.60

POSITIVE LOGITS

 doctor

0.91

 prescribing

0.84

doctor

0.84

 diagn

0.83

icians

0.82

 doctors

0.82

 examiner

0.82

aret

0.82

 care

0.81

 diagnosed

0.80

Activations Density 0.017%

phrases related to medical professionals, particularly doctors

instances of the word "doctors" and related medical terminology

No Comments

No Known Activations