INDEX

Explanations

mentions of specific names or entities

oai_token-act-pair · gpt-3.5-turbo

instances of indirect references or symbols related to specific entities or names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.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

lda

-0.74

 parity

-0.67

 Kenyan

-0.66

iliar

-0.64

 scatter

-0.64

 condol

-0.64

 tremend

-0.63

 dust

-0.63

ioned

-0.62

 scattering

-0.61

POSITIVE LOGITS

£

0.99

¹

0.98

âĹ¼

0.88

ħĭ

0.84

ı

0.83

İ

0.83

į

0.83

Į

0.82

¬

0.81

Phill

0.80

Activations Density 0.349%

mentions of specific names or entities

instances of indirect references or symbols related to specific entities or names

No Comments

No Known Activations