INDEX

Explanations

terms related to immunology, such as immunization and immunotherapy

oai_token-act-pair · gpt-3.5-turbo

terms related to immunology

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

\\\\\\\\

-0.88

Else

-0.83

 Pound

-0.77

lihood

-0.74

BOOK

-0.71

 Witches

-0.71

ÙĲ

-0.70

holder

-0.70

 Shelby

-0.69

 halfway

-0.69

POSITIVE LOGITS

icip

1.09

 immun

1.06

ogl

1.05

obl

1.00

alling

0.97

ophobic

0.95

archs

0.94

icide

0.93

ities

0.93

itions

0.91

Activations Density 0.013%

terms related to immunology, such as immunization and immunotherapy

terms related to immunology

No Comments

No Known Activations