INDEX

Explanations

terms related to medical conditions, specifically cancer

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

stanbul

-0.82

BOOK

-0.81

assy

-0.77

*/(

-0.75

lay

-0.71

theless

-0.71

 Polk

-0.71

é¾įåĸļå£«

-0.70

worldly

-0.70

witz

-0.68

POSITIVE LOGITS

ous

1.35

 chemotherapy

1.14

 diagnosis

1.12

 cancer

1.06

 metast

1.06

 tumors

1.04

 screenings

1.01

 diagnoses

1.00

oma

0.98

 patients

0.97

Activations Density 0.030%

terms related to medical conditions, specifically cancer

No Comments

No Known Activations