INDEX

Explanations

entities or people names

oai_token-act-pair · gpt-3.5-turbo

names of prominent figures or entities

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

pmwiki

-0.69

Enlarge

-0.68

lihood

-0.68

*.

-0.67

 paranormal

-0.66

Motor

-0.64

rawdownloadcloneembedreportprint

-0.64

flies

-0.63

[&

-0.62

 Thumbnails

-0.61

POSITIVE LOGITS

uterte

0.93

addafi

0.90

 Jindal

0.88

 Obama

0.84

 Modi

0.84

Obama

0.84

 Hollande

0.81

DonaldTrump

0.80

 Macron

0.79

 Erdogan

0.79

Activations Density 0.303%

entities or people names

names of prominent figures or entities

No Comments

No Known Activations