INDEX

Explanations

questions starting with "Where" and "does"

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

asts

-0.74

quel

-0.71

apt

-0.69

)].

-0.68

emed

-0.64

ATURE

-0.63

astics

-0.62

 Examiner

-0.61

efe

-0.60

ems

-0.59

POSITIVE LOGITS

abouts

1.41

upon

1.28

fore

1.20

ver

0.96

 else

0.92

with

0.76

soever

0.72

 Else

0.69

velt

0.68

ħ

0.68

Activations Density 0.338%

questions starting with "Where" and "does"

No Comments

No Known Activations