INDEX

Explanations

phrases related to the continuity or maintenance of specific plans or policies

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atana

-0.67

GOT

-0.66

»Ĵ

-0.65

azo

-0.65

aster

-0.64

ongyang

-0.64

shapeshifter

-0.63

¶ħ

-0.63

PsyNetMessage

-0.62

arted

-0.61

POSITIVE LOGITS

 indefinitely

0.90

 uninterrupted

0.89

 unchanged

0.83

 until

0.79

 intact

0.78

 forever

0.76

igion

0.70

 exist

0.67

 awhile

0.66

 afloat

0.66

Activations Density 0.165%

phrases related to the continuity or maintenance of specific plans or policies

No Comments

No Known Activations