INDEX

Explanations

political terms and references

oai_token-act-pair · gpt-3.5-turbo

discussion of political events and their implications

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

[/

-0.67

aeper

-0.65

Kids

-0.65

rawl

-0.64

ertation

-0.64

 Registry

-0.63

Liter

-0.63

cyl

-0.63

 genders

-0.62

Anim

-0.62

POSITIVE LOGITS

 embold

1.05

Putin

1.04

 Kremlin

1.02

Moscow

1.02

 prag

1.00

 Moscow

0.98

 populist

0.98

 geop

0.98

 Lavrov

0.97

 Putin

0.96

Activations Density 1.123%

political terms and references

discussion of political events and their implications

No Comments

No Known Activations