INDEX

Explanations

the presence of the letter "P" in various forms, indicating its focus on proper nouns or specific terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

ģĸ

-0.75

adm

-0.74

 Aval

-0.70

Pry

-0.69

 susp

-0.69

wip

-0.68

 phyl

-0.68

 jewels

-0.68

 diplom

-0.66

 differe

-0.66

POSITIVE LOGITS

redict

1.38

ossible

1.38

ossession

1.28

aired

1.25

ossibly

1.22

ractical

1.22

overty

1.21

rotein

1.19

ardon

1.18

ivot

1.18

Activations Density 0.031%

the presence of the letter "P" in various forms, indicating its focus on proper nouns or specific terms

No Comments

No Known Activations