INDEX

Explanations

phrases indicating a specific sequence or order of steps

oai_token-act-pair · gpt-3.5-turbo

references to instructions or guidelines

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

minist

-0.71

izens

-0.68

aukee

-0.65

ahon

-0.64

utters

-0.63

ollar

-0.62

checked

-0.62

vere

-0.61

irts

-0.61

 destro

-0.60

POSITIVE LOGITS

:-

1.04

:{

0.95

0.94

:"

0.90

:[

0.89

:]

0.88

0.87

):

0.85

>:

0.83

}:

0.83

Activations Density 0.041%

phrases indicating a specific sequence or order of steps

references to instructions or guidelines

No Comments

No Known Activations