INDEX

Explanations

occurrences of different code commands, language syntax, and programming-related terms

oai_token-act-pair · gpt-3.5-turbo

punctuation marks, specifically periods and colons

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 affili

-0.78

antz

-0.77

 disemb

-0.70

 waged

-0.70

 condol

-0.67

ierrez

-0.66

 taboo

-0.66

 foul

-0.66

 mourn

-0.66

nell

-0.65

POSITIVE LOGITS

 Alternatively

1.37

 Therefore

1.32

 However

1.31

 Suppose

1.30

 Additionally

1.30

 Specifically

1.29

 Hence

1.28

 Fortunately

1.26

 Using

1.25

 Luckily

1.25

Activations Density 0.252%

occurrences of different code commands, language syntax, and programming-related terms

punctuation marks, specifically periods and colons

No Comments

No Known Activations