INDEX

Explanations

phrases related to incidents where individuals are taken to hospitals or medical centers

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the transfer of individuals to hospitals following incidents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

sylv

-0.80

 intermitt

-0.72

 striving

-0.71

 desires

-0.65

 enlarg

-0.64

 minded

-0.64

laz

-0.64

 nostalg

-0.63

 thirst

-0.63

 eman

-0.63

POSITIVE LOGITS

 hospital

1.18

ospital

1.02

 Hospital

0.89

 Bellev

0.89

 jail

0.87

 prison

0.83

hospital

0.82

apego

0.82

 Advocate

0.81

 Westminster

0.80

Activations Density 0.093%

phrases related to incidents where individuals are taken to hospitals or medical centers

phrases indicating the transfer of individuals to hospitals following incidents

No Comments

No Known Activations