INDEX

Explanations

phrases related to criticism or evaluation of 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

 Remem

-0.70

 hatch

-0.70

GoldMagikarp

-0.70

 Roses

-0.69

 Caribbean

-0.69

 Citation

-0.68

 Carbuncle

-0.66

Pik

-0.64

 Presbyter

-0.63

 Nemesis

-0.63

POSITIVE LOGITS

idas

1.26

hoc

1.26

roit

1.22

elaide

1.21

ieu

1.20

obe

1.17

vertising

1.15

hesion

1.13

verts

1.13

vertis

1.10

Activations Density 0.017%

phrases related to criticism or evaluation of actions

No Comments

No Known Activations