INDEX

Explanations

prepositions followed by a noun phrase

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

=]

-0.68

CLASSIFIED

-0.60

cia

-0.59

ŀ

-0.56

HRC

-0.56

lain

-0.55

WOM

-0.55

ľ

-0.54

NOW

-0.53

STER

-0.53

POSITIVE LOGITS

ordinate

1.21

 lieu

1.15

 unison

1.11

between

1.11

escap

1.08

animate

1.05

humane

1.03

 accordance

1.02

wards

1.02

clusions

1.00

Activations Density 0.625%

prepositions followed by a noun phrase

No Comments

No Known Activations