INDEX

Explanations

legal and copyright-related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to legal concepts and procedures

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

anmar

-0.76

kefeller

-0.75

ggles

-0.72

itsch

-0.69

zbollah

-0.62

nir

-0.61

ï¸ı

-0.61

tons

-0.60

 lows

-0.60

lez

-0.60

POSITIVE LOGITS

 pursuant

1.00

 lawfully

0.95

 thereof

0.94

 incidental

0.87

 Applicant

0.85

 thereto

0.84

.''.

0.83

or

0.82

 evid

0.78

 ineligible

0.78

Activations Density 1.129%

legal and copyright-related terms

terms related to legal concepts and procedures

No Comments

No Known Activations