INDEX

Explanations

terms related to maximizing efficiency or outcomes

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

NetMessage

-0.79

MORE

-0.74

BuyableInstoreAndOnline

-0.73

bane

-0.70

 Strait

-0.69

 Kimber

-0.67

fixes

-0.67

 disbel

-0.65

Beck

-0.65

 Canary

-0.61

POSITIVE LOGITS

imum

1.21

illian

1.17

icum

1.09

ized

1.07

ising

1.01

izable

0.96

imize

0.96

izes

0.96

ised

0.96

iversal

0.91

Activations Density 0.006%

terms related to maximizing efficiency or outcomes

No Comments

No Known Activations