INDEX

Explanations

lessons learned or knowledge acquired

oai_token-act-pair · gpt-3.5-turbo

phrases related to learning and gaining knowledge

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lished

-0.67

chairs

-0.63

epad

-0.60

ItemImage

-0.59

endez

-0.59

 backdrop

-0.58

 è£ıè¦ļéĨĴ

-0.57

 Fees

-0.57

icted

-0.57

forth

-0.57

POSITIVE LOGITS

 lesson

1.58

 lessons

1.37

 valuable

1.05

 invaluable

1.05

 Lessons

1.03

 firsthand

0.99

 tricks

0.98

 Learned

0.94

 learned

0.94

 learnt

0.92

Activations Density 0.180%

lessons learned or knowledge acquired

phrases related to learning and gaining knowledge

No Comments

No Known Activations