INDEX

Explanations

instances where permission, passes, or clearance is granted

oai_token-act-pair · gpt-3.5-turbo

terms related to permissions, passes, and authorizations

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

unct

-0.74

 corrid

-0.71

 nostalg

-0.70

 Tycoon

-0.70

 rall

-0.69

 fireplace

-0.66

ronics

-0.65

 colle

-0.65

Peb

-0.63

 Bulg

-0.62

POSITIVE LOGITS

 waivers

1.01

 permission

1.01

 waiver

0.97

 permitting

0.91

 visa

0.91

license

0.89

 authorization

0.88

 permits

0.87

 granted

0.84

 permit

0.84

Activations Density 0.336%

instances where permission, passes, or clearance is granted

terms related to permissions, passes, and authorizations

No Comments

No Known Activations