INDEX

Explanations

locations or addresses followed by descriptions or events associated with that location

oai_token-act-pair · gpt-3.5-turbo

punctuation marks and their occurrences within the document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ocalypse

-0.81

²¾

-0.74

algia

-0.71

ãĥīãĥ©

-0.67

phas

-0.65

utorial

-0.62

erous

-0.62

ei

-0.62

ende

-0.61

ivia

-0.61

POSITIVE LOGITS

 located

1.31

 adjacent

1.20

 overlooking

1.18

 situated

1.18

 where

1.15

near

1.12

 adjoining

1.04

 near

1.04

where

1.04

 whence

1.03

Activations Density 0.225%

locations or addresses followed by descriptions or events associated with that location

punctuation marks and their occurrences within the document

No Comments

No Known Activations