INDEX

Explanations

proper nouns and locations

oai_token-act-pair · gpt-3.5-turbo

specific punctuation and formatting elements, as well as indicators of dialogue and names

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

 race

-0.54

 gratification

-0.54

 races

-0.53

ARE

-0.51

 payoff

-0.51

rpm

-0.51

 descriptor

-0.50

 watchdog

-0.50

INTON

-0.50

 Dickinson

-0.49

POSITIVE LOGITS

soDeliveryDate

0.95

Downloadha

0.88

Äĩ

0.80

 Torment

0.68

ymes

0.65

gmail

0.63

 Unified

0.62

sic

0.62

ãĥ¼ãĥ«

0.61

 Parish

0.61

Activations Density 0.960%

proper nouns and locations

specific punctuation and formatting elements, as well as indicators of dialogue and names

No Comments

No Known Activations