INDEX

Explanations

words related to medical diseases or health conditions

oai_token-act-pair · gpt-3.5-turbo

references to diseases and related health conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ramid

-0.91

assy

-0.87

)=(

-0.74

ilts

-0.72

raltar

-0.71

amera

-0.71

adesh

-0.70

lus

-0.70

eele

-0.67

attery

-0.66

POSITIVE LOGITS

 outbreak

1.23

 outbreaks

1.15

 Outbreak

1.08

 diagnosis

1.07

 plag

1.02

 susceptibility

1.02

 epidemic

1.00

 prevalence

0.99

 severity

0.99

 symptoms

0.98

Activations Density 0.053%

words related to medical diseases or health conditions

references to diseases and related health conditions

No Comments

No Known Activations