INDEX

Explanations

mentions of medical patients and their care

oai_token-act-pair · gpt-3.5-turbo

references to patients in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ģ«

-0.73

GGGG

-0.70

Politics

-0.65

OPA

-0.65

angular

-0.63

FLAG

-0.62

shell

-0.62

AUT

-0.62

POL

-0.61

ãĥ¼ãĥĨãĤ£

-0.61

POSITIVE LOGITS

 patients

1.07

ients

1.03

 Patients

0.90

 diagnosed

0.84

 patient

0.80

ysis

0.80

 hospitalized

0.79

 confidentiality

0.75

ubs

0.74

iology

0.73

Activations Density 0.014%

mentions of medical patients and their care

references to patients in various contexts

No Comments

No Known Activations