INDEX

Explanations

instances of legal charges or accusations

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

 livest

-0.67

 obser

-0.66

angular

-0.65

VIS

-0.64

 Remastered

-0.64

atters

-0.63

ullivan

-0.63

 Orth

-0.62

Begin

-0.62

Ħ¢

-0.61

POSITIVE LOGITS

heet

1.24

 criminally

1.02

llah

0.98

 charges

0.91

 indict

0.83

illo

0.82

 indicted

0.80

 indictment

0.76

 accused

0.72

 arra

0.72

Activations Density 0.017%

instances of legal charges or accusations

No Comments

No Known Activations