INDEX

Explanations

quotes or reported speech in an interview setting

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĪĴ

-0.80

 favorably

-0.70

reetings

-0.69

¥µ

-0.68

lihood

-0.67

 favoring

-0.67

uga

-0.64

ĻĤ

-0.63

½

-0.62

 utilizing

-0.62

POSITIVE LOGITS

Scroll

1.21

Labour

0.94

 Trafford

0.89

Scotland

0.81

Shape

0.79

 mould

0.79

alys

0.75

 Ukip

0.74

RAF

0.74

Coun

0.73

Activations Density 1.534%

quotes or reported speech in an interview setting

No Comments

No Known Activations