INDEX

Explanations

keywords related to control points or restrictions, such as gates and gatekeepers

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

ortium

-0.83

 Hots

-0.83

ensional

-0.75

issance

-0.71

ity

-0.69

TING

-0.68

arus

-0.66

ember

-0.65

 Norm

-0.65

ynasty

-0.64

POSITIVE LOGITS

keepers

1.34

keeper

1.28

ways

1.14

keeping

1.01

posts

1.01

fold

0.88

house

0.88

way

0.87

 gates

0.85

stones

0.82

Activations Density 0.018%

keywords related to control points or restrictions, such as gates and gatekeepers

No Comments

No Known Activations