INDEX

Explanations

terms related to policies, administrations, and politics

oai_token-act-pair · gpt-3.5-turbo

references to government actions and associated entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 proble

-0.57

 inval

-0.57

 constitu

-0.55

rul

-0.54

 skelet

-0.54

awaru

-0.51

âĺĨ

-0.51

 defin

-0.50

 Annotations

-0.50

 subconscious

-0.49

POSITIVE LOGITS

lihood

0.73

cipled

0.64

itarian

0.60

isan

0.56

pit

0.55

Enlarge

0.54

 Benghazi

0.53

ÂŃ

0.52

xit

0.51

 Fiscal

0.50

Activations Density 1.918%

terms related to policies, administrations, and politics

references to government actions and associated entities

No Comments

No Known Activations