INDEX

Explanations

items or components that are essential or complete within a certain context

oai_token-act-pair · gpt-3.5-turbo

phrases indicating completeness or possessing everything needed

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ales

-0.74

chard

-0.67

grad

-0.66

gee

-0.63

wagon

-0.62

 Cosponsors

-0.62

leep

-0.62

iatus

-0.61

rame

-0.59

 CrossRef

-0.59

POSITIVE LOGITS

 requisite

1.09

 essentials

1.03

 necessary

1.00

 sorts

0.95

 kinds

0.92

 imaginable

0.89

 necessities

0.86

 amenities

0.84

igator

0.80

 needed

0.78

Activations Density 0.128%

items or components that are essential or complete within a certain context

phrases indicating completeness or possessing everything needed

No Comments

No Known Activations