INDEX

Explanations

proper nouns, especially related to locations and people

oai_token-act-pair · gpt-3.5-turbo

significant entities, such as locations, people, and organizations mentioned in news or reports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 =================================================================

-0.58

================================================================

-0.57

 Canaver

-0.55

Dialogue

-0.54

 Tolkien

-0.53

 Loft

-0.52

 Seah

-0.50

 Historic

-0.50

.).

-0.49

'.

-0.49

POSITIVE LOGITS

omach

0.46

 routed

0.45

ÃĥÃĤ

0.45

emale

0.44

physical

0.44

ecause

0.43

Ö¼

0.43

 physically

0.42

 overpower

0.42

utterstock

0.41

Activations Density 2.210%

proper nouns, especially related to locations and people

significant entities, such as locations, people, and organizations mentioned in news or reports

No Comments

No Known Activations