INDEX

Explanations

phrases related to legal or governmental activities

oai_token-act-pair · gpt-3.5-turbo

the word "In" followed by references to time periods or specific events

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

 unemploy

-0.81

 nurt

-0.68

$$

-0.67

 entertained

-0.65

fab

-0.62

 mathemat

-0.62

 murderer

-0.62

 looms

-0.62

 murd

-0.61

 trave

-0.60

POSITIVE LOGITS

spect

1.31

 addition

1.28

cluding

1.19

ventory

1.18

cluded

1.15

mates

1.08

jured

1.08

 response

1.07

 contrast

1.07

jury

0.98

Activations Density 0.113%

phrases related to legal or governmental activities

the word "In" followed by references to time periods or specific events

No Comments

No Known Activations