INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

instances of the word "An" followed by numerical or significant contextual information

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

BOOK

-0.77

âĸº

-0.77

BLE

-0.72

è¯

-0.72

å¾

-0.68

EngineDebug

-0.68

ðŁĳ

-0.66

 pupils

-0.66

åĨ

-0.65

å¯

-0.64

POSITIVE LOGITS

omaly

1.23

onym

1.19

alogue

1.15

alyses

1.15

notation

1.14

alys

1.13

omal

1.11

terior

1.08

esthetic

1.07

notations

1.06

Activations Density 0.086%

proper nouns

instances of the word "An" followed by numerical or significant contextual information

No Comments

No Known Activations