INDEX

Explanations

terms related to work or tasks within a technical or professional context

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Ukrain

-0.96

wcs

-0.84

iren

-0.74

EStream

-0.69

 Flavoring

-0.68

ylon

-0.67

 Bubble

-0.67

 champagne

-0.67

 constitu

-0.66

 Gord

-0.65

POSITIVE LOGITS

 ethic

1.42

station

1.40

bench

1.36

manship

1.34

flows

1.33

horse

1.31

aday

1.26

forces

1.13

fare

1.06

tops

1.01

Activations Density 0.054%

terms related to work or tasks within a technical or professional context

No Comments

No Known Activations