INDEX

Explanations

phrases related to medical procedures, hospitals, and healthcare settings

oai_token-act-pair · gpt-3.5-turbo

references to medical facilities and healthcare-related activities

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

 Flavoring

-0.82

uble

-0.81

xtap

-0.72

 Strongh

-0.71

 newsp

-0.68

 Stronghold

-0.67

wcsstore

-0.66

 Saur

-0.66

isSpecialOrderable

-0.64

htaking

-0.64

POSITIVE LOGITS

iatrics

1.09

 physicians

1.09

 patients

1.07

iatric

1.06

 paramed

1.06

ysis

1.06

 hospital

1.06

 doctor

1.05

 physician

1.04

 practitioner

1.00

Activations Density 0.148%

phrases related to medical procedures, hospitals, and healthcare settings

references to medical facilities and healthcare-related activities

No Comments

No Known Activations