INDEX

Explanations

phrases related to physical interaction or actions

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

enegger

-0.77

picture

-0.64

 Thrones

-0.64

 EDITION

-0.63

OME

-0.63

 Pulse

-0.62

Kul

-0.62

 Valhalla

-0.62

INGTON

-0.61

 Judgment

-0.61

POSITIVE LOGITS

ggy

1.19

eps

1.11

gging

1.11

eking

1.10

eper

1.05

achy

1.03

pperc

1.03

eping

1.02

formance

1.02

asant

0.98

Activations Density 0.015%

phrases related to physical interaction or actions

No Comments

No Known Activations