INDEX

Explanations

phrases or words indicating broad and unrestricted spatial scope

oai_token-act-pair · gpt-3.5-turbo

phrases related to versatility and limitless possibilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

endas

-0.67

orman

-0.65

Rue

-0.65

Wa

-0.64

utz

-0.64

alus

-0.63

 subtitles

-0.62

illard

-0.62

alion

-0.61

uers

-0.61

POSITIVE LOGITS

 imaginable

1.53

 conceivable

1.11

THING

0.99

 else

0.83

DragonMagazine

0.80

Else

0.80

where

0.76

 possible

0.74

body

0.73

 practicable

0.70

Activations Density 0.107%

phrases or words indicating broad and unrestricted spatial scope

phrases related to versatility and limitless possibilities

No Comments

No Known Activations