INDEX

Explanations

terms related to news, events, and actions of political figures

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate mental health issues and behavior patterns

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

âĢĳ

-1.04

lean

-0.86

 nineteen

-0.82

 glim

-0.80

Newsletter

-0.77

 eighteen

-0.75

 eighty

-0.72

iHUD

-0.71

etheless

-0.70

 Canaver

-0.69

POSITIVE LOGITS

1.84

1.54

pic

1.37

.#

1.36

 https

1.25

1.24

 ðŁĳ

1.22

ðŁĺ

1.22

 ðŁ

1.21

ðŁ

1.21

Activations Density 0.855%

terms related to news, events, and actions of political figures

phrases that indicate mental health issues and behavior patterns

No Comments

No Known Activations