INDEX

Explanations

words related to continuation, ongoing actions, or persistence

oai_token-act-pair · gpt-3.5-turbo

negative phrases and expressions of doubt or uncertainty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Jav

-0.68

Ha

-0.67

è¦ļéĨĴ

-0.67

 guid

-0.66

 Advice

-0.66

Ki

-0.65

 Advertisement

-0.65

iT

-0.64

Mek

-0.63

Ens

-0.62

POSITIVE LOGITS

 adolesc

0.82

}}}

0.73

arnaev

0.73

ORED

0.71

bilt

0.69

olina

0.67

ledged

0.66

clusively

0.65

ued

0.65

 untouched

0.65

Activations Density 0.241%

words related to continuation, ongoing actions, or persistence

negative phrases and expressions of doubt or uncertainty

No Comments

No Known Activations