INDEX

Explanations

phrases related to negotiation, consent, and relationships

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

 timely

-0.76

 boarded

-0.74

 affili

-0.73

 penned

-0.73

ciplinary

-0.72

uly

-0.71

 surname

-0.71

 broadly

-0.70

 attire

-0.69

 brief

-0.69

POSITIVE LOGITS

 Eventually

1.47

 Then

1.32

 Sometimes

1.29

Eventually

1.25

 Needless

1.24

 Usually

1.24

 Finally

1.23

 Sounds

1.20

 Afterwards

1.19

 Occasionally

1.17

Activations Density 0.388%

phrases related to negotiation, consent, and relationships

No Comments

No Known Activations