INDEX

Explanations

prepositions followed by nouns or verb phrases indicating conflict or opposition

oai_token-act-pair · gpt-3.5-turbo

prepositions and phrases indicating opposition or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

mask

-0.73

âĨĳ

-0.65

liner

-0.65

fortunately

-0.60

LV

-0.59

MU

-0.58

KNOWN

-0.58

vill

-0.58

hent

-0.58

CLS

-0.58

POSITIVE LOGITS

 terms

0.63

 behalf

0.61

 regard

0.59

assian

0.59

tein

0.57

 bidding

0.56

 transact

0.55

 bids

0.55

ertodd

0.54

ussian

0.54

Activations Density 0.586%

prepositions followed by nouns or verb phrases indicating conflict or opposition

prepositions and phrases indicating opposition or conditions

No Comments

No Known Activations