INDEX

Explanations

words related to internal affairs and operations within organizations

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

eday

-0.81

ammy

-0.79

kk

-0.78

apo

-0.76

ky

-0.75

kers

-0.74

 Maker

-0.74

ggies

-0.73

geon

-0.71

Pitt

-0.71

POSITIVE LOGITS

 combustion

1.33

 workings

1.18

ized

0.94

 organs

0.87

 turmoil

0.82

most

0.78

 affairs

0.77

 lobe

0.77

 strife

0.77

izing

0.76

Activations Density 0.568%

words related to internal affairs and operations within organizations

No Comments

No Known Activations