INDEX

Explanations

references to medical care and treatment of patients

oai_token-act-pair · gpt-3.5-turbo

mentions of "patients."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

angular

-0.87

sg

-0.70

ģ«

-0.69

OPA

-0.68

oras

-0.67

GGGG

-0.65

ball

-0.63

ãĥ¼ãĥĨãĤ£

-0.63

shell

-0.62

gers

-0.62

POSITIVE LOGITS

 patients

0.95

ients

0.89

 Patients

0.88

 diagnosed

0.85

 patient

0.83

ysis

0.81

inical

0.80

iology

0.79

 hospitalized

0.79

 confidentiality

0.77

Activations Density 0.036%

references to medical care and treatment of patients

mentions of "patients."

No Comments

No Known Activations