INDEX

Explanations

terms related to healthcare

oai_token-act-pair · gpt-3.5-turbo

references to healthcare concepts and issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ãĤ¨ãĥ«

-0.82

IGN

-0.81

wcsstore

-0.79

urat

-0.77

ãĥĥãĥī

-0.77

KEN

-0.76

 sqor

-0.74

NK

-0.71

iple

-0.69

urb

-0.69

POSITIVE LOGITS

 providers

1.12

giving

1.06

 healthcare

1.01

amacare

0.98

 clinics

0.96

 professionals

0.96

 provider

0.96

 services

0.90

aceutical

0.88

 nurses

0.85

Activations Density 0.028%

terms related to healthcare

references to healthcare concepts and issues

No Comments

No Known Activations