INDEX

Explanations

statements or predictions made by individuals

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 originals

-0.72

addons

-0.64

 disciplines

-0.63

 Administ

-0.63

ynski

-0.63

endars

-0.63

azeera

-0.63

issions

-0.62

 Models

-0.61

 Cosponsors

-0.61

POSITIVE LOGITS

1.12

:'

1.05

 whereby

1.03

:[

1.02

:-

0.97

!:

0.96

:"

0.96

:(

0.94

:#

0.93

:,

0.91

Activations Density 0.266%

statements or predictions made by individuals

No Comments

No Known Activations