INDEX

Explanations

mentions of legal proceedings and law enforcement

oai_token-act-pair · gpt-3.5-turbo

organizational or institutional names, particularly those related to legal or governmental contexts

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

Defense

-0.85

avored

-0.81

 Whitman

-0.80

 favored

-0.76

colored

-0.74

rollment

-0.74

 favors

-0.74

GOP

-0.74

caliber

-0.74

imize

-0.74

POSITIVE LOGITS

 recognise

1.34

 Â£

1.34

 realise

1.31

 centre

1.29

 apologise

1.23

 apologised

1.21

 Ukip

1.20

 aluminium

1.18

 Labour

1.18

 recognised

1.18

Activations Density 3.434%

mentions of legal proceedings and law enforcement

organizational or institutional names, particularly those related to legal or governmental contexts

No Comments

No Known Activations