INDEX

Explanations

The neuron is looking for mentions of political parties and related activities

oai_token-act-pair · gpt-3.5-turbo

mentions of political parties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ravings

-0.80

alam

-0.77

 Grizz

-0.75

aneously

-0.73

Hoo

-0.70

apons

-0.70

obar

-0.66

 Wheat

-0.66

alez

-0.65

 Dull

-0.65

POSITIVE LOGITS

 affiliation

1.29

 affili

1.05

 leadership

1.05

 primaries

0.98

 nominate

0.97

 nominating

0.97

 caucus

0.96

leader

0.95

 nominee

0.95

 caucuses

0.94

Activations Density 0.052%

The neuron is looking for mentions of political parties and related activities

mentions of political parties

No Comments

No Known Activations