INDEX

Explanations

phrases related to medical conditions, specifically focusing on the condition of patients

oai_token-act-pair · gpt-3.5-turbo

phrases related to medical conditions and injury severity

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

soDeliveryDate

-0.82

 monog

-0.80

Publisher

-0.77

 monopol

-0.75

 Agric

-0.70

 Straight

-0.69

 utopian

-0.69

 Clubs

-0.69

Myth

-0.69

opp

-0.68

POSITIVE LOGITS

EMS

0.94

llah

0.93

rapnel

0.87

nsics

0.86

raints

0.82

 injuries

0.81

 decomp

0.79

letal

0.79

nsic

0.78

 recovered

0.77

Activations Density 0.217%

phrases related to medical conditions, specifically focusing on the condition of patients

phrases related to medical conditions and injury severity

No Comments

No Known Activations