INDEX

Explanations

phrases indicating emphasis or importance

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Def" and its variations, indicating definitions or defined terms

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sth

-0.80

 Dragonbound

-0.71

ï¸

-0.68

Reviewer

-0.67

natureconservancy

-0.66

mares

-0.65

 Holly

-0.64

)=(

-0.62

worldly

-0.62

 Boll

-0.62

POSITIVE LOGITS

initions

1.36

initely

1.29

erred

1.25

icient

1.21

endants

1.21

ence

1.19

ences

1.16

inite

1.11

endant

1.10

erence

1.10

Activations Density 0.023%

phrases indicating emphasis or importance

instances of the word "Def" and its variations, indicating definitions or defined terms

No Comments

No Known Activations