INDEX

Explanations

special characters and non-English words

oai_token-act-pair · gpt-3.5-turbo

special characters and symbols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

orius

-0.75

 Sorceress

-0.66

 CLSID

-0.63

 incoming

-0.63

 frontal

-0.62

Oro

-0.62

 detailed

-0.62

eur

-0.61

 voic

-0.61

 critically

-0.60

POSITIVE LOGITS

¶

1.13

³

0.95

º

0.92

éŃĶ

0.92

Ð´

0.90

ï¸

0.89

LOG

0.89

½

0.85

iece

0.85

ãĤ£

0.85

Activations Density 0.005%

special characters and non-English words

special characters and symbols

No Comments

No Known Activations