INDEX

Explanations

phrases related to legal or criminal activities

oai_token-act-pair · gpt-3.5-turbo

punctuation marks, specifically commas, within the text

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ļéĨĴ

-0.75

orem

-0.74

¬¼

-0.72

 Dragonbound

-0.72

amus

-0.72

ukong

-0.67

myra

-0.66

KEN

-0.66

glas

-0.65

iru

-0.65

POSITIVE LOGITS

 including

1.50

 ranging

1.47

including

1.27

 includ

1.25

 totaling

1.25

 each

1.21

 spanning

1.14

all

1.11

 Including

1.07

 none

1.06

Activations Density 0.379%

phrases related to legal or criminal activities

punctuation marks, specifically commas, within the text

No Comments

No Known Activations