INDEX

Explanations

characters that are not part of standard English text

oai_token-act-pair · gpt-3.5-turbo

special characters or non-standard symbols in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

alion

-0.76

icter

-0.72

 Norn

-0.71

iple

-0.68

aimon

-0.67

 plaque

-0.65

apers

-0.65

atre

-0.65

ulhu

-0.65

ified

-0.64

POSITIVE LOGITS

Â¢

1.04

ÃĤ

1.00

Â©

0.95

âĢ¢âĢ¢

0.89

Â£

0.83

âĤ¬

0.81

Â·Â·

0.80

Johnson

0.79

Â¯Â¯

0.79

Â§

0.77

Activations Density 0.008%

characters that are not part of standard English text

special characters or non-standard symbols in the text

No Comments

No Known Activations