INDEX

Explanations

specific instances where an action is either allowed or not allowed

oai_token-act-pair · gpt-3.5-turbo

words related to permission and authorization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

tal

-0.63

stats

-0.63

guy

-0.63

 bust

-0.63

box

-0.62

pop

-0.61

beat

-0.60

Voc

-0.60

 tons

-0.60

ðŁ

-0.60

POSITIVE LOGITS

 permitted

3.27

 permissible

2.30

 prohibited

2.00

 allowable

1.85

 authorized

1.73

 allowed

1.68

 forbidden

1.61

 permit

1.60

 permitting

1.56

 authorised

1.54

Activations Density 0.011%

specific instances where an action is either allowed or not allowed

words related to permission and authorization

No Comments

No Known Activations