INDEX

Explanations

updates or new information in various contexts such as news articles and reports

oai_token-act-pair · gpt-3.5-turbo

words related to updates and postings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Till

-0.66

Newsletter

-0.60

ãĤ¼ãĤ¦ãĤ¹

-0.60

esm

-0.59

 Sakura

-0.58

teenth

-0.56

ãĥį

-0.56

cuts

-0.56

 neighboring

-0.55

stood

-0.54

POSITIVE LOGITS

Updated

0.66

hemy

0.63

↵

0.62

acia

0.62

elaide

0.60

inet

0.59

inion

0.58

}}

0.58

 Chronicle

0.57

ir

0.56

Activations Density 0.021%

updates or new information in various contexts such as news articles and reports

words related to updates and postings

No Comments

No Known Activations