INDEX

Explanations

mentions of medical terms and entities

oai_token-act-pair · gpt-3.5-turbo

terms related to medical topics and health conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

chnology

-0.72

ãĤ©

-0.68

holding

-0.68

holders

-0.67

ordinate

-0.62

resy

-0.60

redits

-0.60

ngth

-0.59

 Grayson

-0.59

holder

-0.58

POSITIVE LOGITS

ieval

1.34

iterranean

1.15

ication

1.04

ician

1.00

ications

0.99

ford

0.94

itations

0.92

itation

0.92

ient

0.91

usa

0.89

Activations Density 0.027%

mentions of medical terms and entities

terms related to medical topics and health conditions

No Comments

No Known Activations