INDEX

Explanations

prepositions followed by the word 'in'

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

idays

-0.68

 incumb

-0.67

killed

-0.63

 bloggers

-0.63

erick

-0.62

onym

-0.62

llor

-0.62

rolet

-0.62

 PLEASE

-0.59

 endors

-0.58

POSITIVE LOGITS

 unison

1.31

 front

1.20

 lieu

1.18

animate

1.16

 accordance

1.15

between

1.13

versions

1.06

 spite

0.99

humane

0.97

escap

0.97

Activations Density 0.339%

prepositions followed by the word 'in'

No Comments

No Known Activations