INDEX

Explanations

legal or judicial terms related to regulations or laws

oai_token-act-pair · gpt-3.5-turbo

references to legal terms and regulations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Kinnikuman

-0.88

lymp

-0.79

romeda

-0.77

 Tycoon

-0.76

upon

-0.76

stadt

-0.73

Ĥ¬

-0.73

hoff

-0.72

eon

-0.72

 Royale

-0.72

POSITIVE LOGITS

ised

0.85

abulary

0.80

 mandate

0.79

otine

0.77

 enforcement

0.76

 authority

0.75

 deadline

0.75

 authorization

0.74

 statutory

0.72

 authorized

0.70

Activations Density 0.005%

legal or judicial terms related to regulations or laws

references to legal terms and regulations

No Comments

No Known Activations