INDEX

Explanations

first-person pronouns in possessive form

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

conom

-0.81

cum

-0.75

puff

-0.75

icter

-0.73

ppings

-0.73

bender

-0.71

tar

-0.70

endra

-0.68

 preceded

-0.68

ussen

-0.68

POSITIVE LOGITS

selves

1.23

own

1.10

 beloved

0.94

 ourselves

0.89

 respective

0.88

 motto

0.88

 ancestors

0.87

 collective

0.87

 asses

0.85

 dear

0.85

Activations Density 0.130%

first-person pronouns in possessive form

No Comments

No Known Activations