INDEX

Explanations

news headlines containing locations and police-related terms

oai_token-act-pair · gpt-3.5-turbo

a series of numerical and temporal indicators related to events or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 renegoti

-0.94

hed

-0.89

 appoint

-0.87

 favour

-0.86

 persuasion

-0.86

 sustainability

-0.85

 veto

-0.85

 mull

-0.82

 optim

-0.81

 endorsing

-0.81

POSITIVE LOGITS

Authorities

1.49

Investigators

1.49

Police

1.41

According

1.38

Copyright

1.30

Officers

1.29

Sources

1.29

Officials

1.29

Around

1.28

RELATED

1.26

Activations Density 0.412%

news headlines containing locations and police-related terms

a series of numerical and temporal indicators related to events or actions

No Comments

No Known Activations