INDEX

Explanations

terms related to a specific standard or norm

oai_token-act-pair · gpt-3.5-turbo

mentions of "standard" across various contexts

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

lez

-0.78

cig

-0.75

Stars

-0.73

resent

-0.69

acan

-0.67

Lens

-0.64

der

-0.64

 destro

-0.63

les

-0.63

rak

-0.63

POSITIVE LOGITS

 deviation

1.53

 deviations

1.41

 bearer

1.16

ised

1.06

ization

0.95

arily

0.92

isation

0.91

izes

0.82

é¾įå¥ĳå£«

0.80

izations

0.78

Activations Density 0.017%

terms related to a specific standard or norm

mentions of "standard" across various contexts

No Comments

No Known Activations