INDEX

Explanations

proper nouns and numbers

oai_token-act-pair · gpt-3.5-turbo

references to significant events or people

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

icho

-0.90

Mich

-0.81

aign

-0.80

 Asuka

-0.75

Gaza

-0.75

ibel

-0.75

icka

-0.74

ãĤ¡

-0.71

inks

-0.71

 IEEE

-0.70

POSITIVE LOGITS

Tr

2.40

Tr

2.16

TR

2.07

tr

2.00

tr

1.95

TR

1.68

Tre

1.49

Tra

1.41

Tri

1.34

Tru

1.34

Activations Density 0.201%

proper nouns and numbers

references to significant events or people

No Comments

No Known Activations