INDEX

Explanations

proper nouns related to people and organizations who write letters

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "letter."

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

 Tune

-0.68

orsi

-0.65

 Occupations

-0.63

illon

-0.63

rans

-0.62

rolet

-0.61

 CCTV

-0.60

 Goods

-0.60

tics

-0.60

negie

-0.60

POSITIVE LOGITS

Letter

0.98

 letter

0.94

velop

0.88

 addressed

0.86

letter

0.85

 letters

0.85

 penned

0.83

 inbox

0.82

press

0.80

boxing

0.79

Activations Density 0.033%

proper nouns related to people and organizations who write letters

occurrences of the word "letter."

No Comments

No Known Activations