INDEX

Explanations

terms related to legal violations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating legal violations or infringements

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

ritz

-0.75

onna

-0.73

ppa

-0.72

retty

-0.69

 dwind

-0.65

 thanking

-0.64

borgh

-0.64

achine

-0.63

agne

-0.62

bang

-0.62

POSITIVE LOGITS

Ö¼

0.91

 norms

0.82

NRS

0.77

 instr

0.76

 Contracts

0.75

orius

0.75

hibited

0.73

 procedural

0.72

CLS

0.70

ãģį

0.70

Activations Density 0.098%

terms related to legal violations

phrases indicating legal violations or infringements

No Comments

No Known Activations