INDEX

Explanations

words related to vegetation, particularly shrubs

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

querque

-0.75

 Dragonbound

-0.74

eering

-0.72

 Gutenberg

-0.67

 politics

-0.64

eers

-0.63

eer

-0.63

 partial

-0.63

FactoryReloaded

-0.62

 largeDownload

-0.62

POSITIVE LOGITS

ines

1.11

anch

1.09

inks

1.09

ifts

1.09

ank

1.08

ith

1.07

ubs

1.06

ifting

1.04

pling

1.00

agging

0.99

Activations Density 0.018%

words related to vegetation, particularly shrubs

No Comments

No Known Activations