INDEX

Explanations

keywords related to hospitals

oai_token-act-pair · gpt-3.5-turbo

mentions of hospitals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

AUT

-0.67

;;;;;;;;

-0.67

 Stronghold

-0.64

lihood

-0.64

---------

-0.63

PU

-0.61

âĢ¢âĢ¢âĢ¢âĢ¢

-0.61

uten

-0.61

aphael

-0.60

 Political

-0.60

POSITIVE LOGITS

itals

1.09

 hospitals

1.02

 traged

0.93

ilitating

0.90

osponsors

0.90

 volunte

0.87

 pharmacies

0.87

 mosqu

0.86

©¶æ

0.84

irtual

0.82

Activations Density 0.010%

keywords related to hospitals

mentions of hospitals

No Comments

No Known Activations