INDEX

Explanations

time-related terms such as "now," "year," and "minutes."

oai_token-act-pair · gpt-3.5-turbo

words indicating approximation or frequency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Difference

-0.62

 Distribut

-0.57

igans

-0.57

 Fabric

-0.56

 Haku

-0.55

oured

-0.54

CHR

-0.53

Saw

-0.53

 Pant

-0.53

pires

-0.51

POSITIVE LOGITS

ilaterally

1.03

terday

1.01

 spor

0.94

 exting

0.88

 peacefully

0.86

ãĤ´ãĥ³

0.86

onto

0.82

 differently

0.81

 orally

0.81

 inconsist

0.80

Activations Density 0.199%

time-related terms such as "now," "year," and "minutes."

words indicating approximation or frequency

No Comments

No Known Activations