INDEX

Explanations

references to quantities or degrees of significance

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

 Tags

-0.89

pta

-0.78

 Releases

-0.74

ATURES

-0.73

 Accessories

-0.71

 Provision

-0.70

ysc

-0.68

 Thieves

-0.68

ocene

-0.68

 Picks

-0.67

POSITIVE LOGITS

 simpler

0.99

 appreciated

0.96

 nicer

0.96

 bigger

0.95

 larger

0.94

 smaller

0.93

rier

0.90

 awaited

0.89

 publicized

0.87

 broader

0.87

Activations Density 0.021%

references to quantities or degrees of significance

No Comments

No Known Activations