INDEX

Explanations

technical terms related to technology and computer science

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 streams

-0.61

 calves

-0.60

 Rover

-0.60

assetsadobe

-0.59

atinum

-0.58

 narrator

-0.58

Wem

-0.57

TPS

-0.57

 Emerson

-0.56

 Penalty

-0.56

POSITIVE LOGITS

venient

1.26

crete

1.20

ventional

1.19

clave

1.18

ventions

1.17

naissance

1.16

structed

1.14

vict

1.14

stant

1.13

undrum

1.11

Activations Density 0.613%

technical terms related to technology and computer science

No Comments

No Known Activations