INDEX

Explanations

the name "Herrera"

oai_token-act-pair · gpt-3.5-turbo

individual names, particularly those related to sports and media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Independence

-0.76

ãĥ´ãĤ¡

-0.73

Origin

-0.72

 Mongo

-0.72

Heat

-0.71

 Islanders

-0.71

¶æ

-0.68

Mask

-0.68

 Madison

-0.68

 Buffer

-0.67

POSITIVE LOGITS

naires

1.02

istry

0.94

brow

0.88

naire

0.87

icide

0.84

igans

0.80

esan

0.79

nered

0.78

osc

0.77

nect

0.76

Activations Density 0.020%

the name "Herrera"

individual names, particularly those related to sports and media

No Comments

No Known Activations