INDEX

Explanations

miscellaneous tips, tools, and information

oai_token-act-pair · gpt-3.5-turbo

punctuation usage and sentence endings

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 gunfire

-0.67

 recess

-0.67

 biological

-0.64

 contraction

-0.64

lier

-0.62

 contingency

-0.62

 alley

-0.62

 coupling

-0.62

 burial

-0.61

 runoff

-0.61

POSITIVE LOGITS

 Includes

1.32

 Including

1.19

 Please

1.07

 Especially

1.06

 Whether

1.05

 Featuring

1.05

 Learn

1.04

 Besides

1.01

Includes

1.00

 Explore

0.96

Activations Density 0.297%

miscellaneous tips, tools, and information

punctuation usage and sentence endings

No Comments

No Known Activations