INDEX

Explanations

terms related to law enforcement and crime investigations, particularly involving shootings and police actions

oai_token-act-pair · gpt-3.5-turbo

references to law enforcement and investigations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

HUD

-0.68

iyah

-0.65

iscons

-0.64

oday

-0.63

aye

-0.62

LET

-0.61

WAS

-0.61

Supported

-0.60

iHUD

-0.60

 Finished

-0.59

POSITIVE LOGITS

 nowadays

1.07

 rarely

0.89

 dictators

0.85

 normally

0.84

 typically

0.83

 seldom

0.83

 notoriously

0.81

 Usually

0.80

 usually

0.80

 tends

0.78

Activations Density 1.998%

terms related to law enforcement and crime investigations, particularly involving shootings and police actions

references to law enforcement and investigations

No Comments

No Known Activations