INDEX

Explanations

phrases related to significant size or scale

oai_token-act-pair · gpt-3.5-turbo

references to significant sizes or quantities

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

roma

-0.81

ilk

-0.80

lest

-0.71

manship

-0.70

say

-0.70

apple

-0.70

nan

-0.70

©¶æ

-0.70

mology

-0.69

Reviewed

-0.69

POSITIVE LOGITS

 amounts

1.30

 amount

1.16

 quantities

1.11

 influx

1.02

 sums

1.00

 swath

0.97

 undertaking

0.96

 earthqu

0.92

 swat

0.91

 proportions

0.88

Activations Density 0.077%

phrases related to significant size or scale

references to significant sizes or quantities

No Comments

No Known Activations