INDEX

Explanations

the name "Tal" followed by a number

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of the name "Tal"

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

EEE

-0.85

lihood

-0.79

vernment

-0.73

âķĲ

-0.71

ãĥĩãĤ£

-0.71

ullivan

-0.69

ï¸

-0.67

CRIP

-0.65

-0.64

 Veronica

-0.64

POSITIVE LOGITS

isman

1.07

iban

0.89

mud

0.87

Tal

0.84

aga

0.82

ented

0.81

ibur

0.80

imony

0.79

ison

0.78

tal

0.77

Activations Density 0.006%

the name "Tal" followed by a number

repeated mentions of the name "Tal"

No Comments

No Known Activations