INDEX

Explanations

phrases related to negative emotions and criticism

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

reen

-0.78

atories

-0.76

cible

-0.71

athering

-0.68

 glim

-0.67

aldi

-0.67

unker

-0.65

oult

-0.64

ativity

-0.63

atory

-0.63

POSITIVE LOGITS

!!!!!

1.30

!!!

1.22

!!

1.18

?!

1.13

!!!!

1.02

!/

0.98

@#

0.98

!"

0.94

??

0.93

@#&

0.91

Activations Density 0.013%

phrases related to negative emotions and criticism

No Comments

No Known Activations