INDEX

Explanations

negative statements and political figures

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

iple

-0.78

utor

-0.68

DragonMagazine

-0.67

ãĥĥãĥĪ

-0.67

 Intake

-0.66

 ILCS

-0.65

 SERV

-0.64

ISE

-0.62

 prest

-0.60

 guiActiveUn

-0.59

POSITIVE LOGITS

 noting

1.35

 saying

1.29

 stating

1.28

 citing

1.13

 insisting

1.07

 mentioning

1.00

 arguing

0.98

 declaring

0.98

 claiming

0.98

 stressing

0.96

Activations Density 0.419%

negative statements and political figures

No Comments

No Known Activations