INDEX

Explanations

references to various colleges and sports teams, particularly in the context of sports events and rankings

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

 pardon

-0.69

orders

-0.68

 smugglers

-0.65

 Luxembourg

-0.65

 compartment

-0.65

è¦ļéĨĴ

-0.64

 Chern

-0.63

ãĤ·

-0.63

arding

-0.62

ãĥĻ

-0.61

POSITIVE LOGITS

 University

1.29

 Athletics

1.27

 athletics

1.26

 athletic

1.17

 Athletic

1.14

 alum

1.13

 freshman

1.10

 alumni

1.05

Football

1.03

 Wildcats

1.02

Activations Density 0.229%

references to various colleges and sports teams, particularly in the context of sports events and rankings

No Comments

No Known Activations