INDEX

Explanations

references to specific groups of people or individuals

oai_token-act-pair · gpt-3.5-turbo

references to different groups of people, particularly women and individuals with medical conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ãĥķãĤ©

-0.90

rawdownloadcloneembedreportprint

-0.81

Assembly

-0.80

Oracle

-0.78

Unity

-0.77

Capital

-0.76

Politics

-0.76

é»Ĵ

-0.75

Platform

-0.75

Shadow

-0.74

POSITIVE LOGITS

opausal

1.10

 diagnosed

1.02

 undergoing

1.00

 ingested

0.98

 aged

0.98

 hormone

0.94

 hospitalized

0.94

 ingest

0.88

olesterol

0.87

 afflicted

0.87

Activations Density 0.178%

references to specific groups of people or individuals

references to different groups of people, particularly women and individuals with medical conditions

No Comments

No Known Activations