INDEX

Explanations

proper nouns related to locations, movies, names, and 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

lehem

-0.75

abase

-0.70

SPONSORED

-0.65

\<

-0.65

atin

-0.62

ortium

-0.61

sole

-0.58

anos

-0.58

ãĤ¼ãĤ¦ãĤ¹

-0.57

Spread

-0.57

POSITIVE LOGITS

kefeller

1.00

backer

0.83

issance

0.77

zl

0.77

restling

0.74

ansom

0.71

 Janeiro

0.69

 Myster

0.67

 Tolkien

0.66

warts

0.65

Activations Density 1.532%

proper nouns related to locations, movies, names, and organizations

No Comments

No Known Activations