INDEX

Explanations

phrases related to personal relationships and conflicts

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

ratulations

-0.75

merce

-0.75

sylv

-0.74

QUI

-0.73

clair

-0.73

TEXT

-0.72

SPONSORED

-0.72

worker

-0.71

Enlarge

-0.70

retty

-0.68

POSITIVE LOGITS

 whence

1.02

 afar

1.00

 thence

0.84

 obscurity

0.78

 orbit

0.70

 captivity

0.66

 scratch

0.66

 Brune

0.65

 existence

0.65

 premises

0.65

Activations Density 0.549%

phrases related to personal relationships and conflicts

No Comments

No Known Activations