INDEX

Explanations

phrases related to evaluations and judgments, particularly in academic or professional contexts

oai_token-act-pair · gpt-3.5-turbo

concepts related to scientific evaluation and publication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

çīĪ

-0.80

luaj

-0.75

ibus

-0.72

BuyableInstoreAndOnline

-0.68

mobi

-0.66

angel

-0.65

ãĤ³

-0.63

ãĤ¸

-0.62

lah

-0.61

DragonMagazine

-0.61

POSITIVE LOGITS

 particular

1.36

 certain

1.36

 something

1.15

 specific

1.09

 given

1.00

 someone

0.96

 individual

0.93

 somebody

0.91

 specified

0.91

 items

0.89

Activations Density 0.647%

phrases related to evaluations and judgments, particularly in academic or professional contexts

concepts related to scientific evaluation and publication

No Comments

No Known Activations