INDEX

Explanations

references to patients and medical records in a hospital setting

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

references to patients and their experiences in medical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

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

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_post

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

oras

-0.78

orem

-0.71

itect

-0.68

Rac

-0.67

 Decoder

-0.66

 Sparrow

-0.66

ãĥ©ãĥ³

-0.66

 Dunk

-0.66

UTF

-0.65

odes

-0.63

POSITIVE LOGITS

 hospitalized

1.10

 undergoing

1.10

 suffering

1.08

 diagnosed

1.01

 patient

0.95

 discharged

0.94

 patients

0.94

 enrolled

0.85

doctor

0.84

 needing

0.82

Activations Density 0.084%

references to patients and medical records in a hospital setting

references to patients and their experiences in medical contexts

No Comments

No Known Activations