INDEX

Explanations

mentions of specific individuals, likely politicians

oai_token-act-pair · gpt-3.5-turbo

key political figures and their affiliations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ANC

-0.81

Gaza

-0.76

 Chelsea

-0.75

MTA

-0.74

 Doodle

-0.73

VID

-0.73

Chelsea

-0.73

 Mubarak

-0.71

 Nepal

-0.70

LTD

-0.69

POSITIVE LOGITS

 Corker

1.07

 McConnell

1.00

 Flake

0.92

 Mitch

0.90

GOP

0.89

 Majority

0.87

 Lindsey

0.84

 Boehner

0.82

 Ways

0.82

 Cruz

0.80

Activations Density 0.253%

mentions of specific individuals, likely politicians

key political figures and their affiliations

No Comments

No Known Activations