INDEX

Explanations

affirmative responses or acknowledgments of a statement

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

idas

-0.72

iry

-0.70

rouse

-0.69

Reviewed

-0.67

mented

-0.66

erd

-0.66

bish

-0.65

aukee

-0.65

Fram

-0.65

idy

-0.62

POSITIVE LOGITS

Ø³

0.73

 there

0.68

 assumes

0.66

Ã±a

0.63

NULL

0.62

 spoil

0.60

adays

0.60

uracy

0.60

 Lanka

0.60

 assuming

0.59

Activations Density 0.344%

affirmative responses or acknowledgments of a statement

No Comments

No Known Activations