INDEX

Explanations

proper nouns related to political figures or organizations

oai_token-act-pair · gpt-3.5-turbo

names of prominent individuals and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

interstitial

-0.60

ãĥ¼ãĥĨãĤ£

-0.53

Els

-0.49

è¦ļéĨĴ

-0.49

ãĥ¼ãĥĨ

-0.48

ä»£

-0.47

 guiName

-0.47

*.

-0.47

asta

-0.47

Cth

-0.46

POSITIVE LOGITS

 could

0.62

 opted

0.62

has

0.60

 spokesman

0.60

 insisted

0.59

 couldn

0.59

 intends

0.59

 responded

0.59

had

0.58

 chose

0.58

Activations Density 0.980%

proper nouns related to political figures or organizations

names of prominent individuals and organizations

No Comments

No Known Activations