INDEX

Explanations

phrases related to physical actions or tasks

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate surprise or astonishment

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

 Syri

-0.78

 Yugoslavia

-0.65

 Kurds

-0.65

 Zionism

-0.64

 Marxism

-0.64

 Daesh

-0.60

entity

-0.59

,[

-0.59

["

-0.58

PKK

-0.58

POSITIVE LOGITS

your

0.75

 carbohyd

0.71

 hobby

0.67

Your

0.66

 beginner

0.66

Kids

0.66

 cute

0.65

 pricey

0.65

 YOUR

0.64

 hassle

0.64

Activations Density 1.807%

phrases related to physical actions or tasks

phrases that indicate surprise or astonishment

No Comments

No Known Activations