INDEX

Explanations

phrases related to the addition or specifying something extra

oai_token-act-pair · gpt-3.5-turbo

references to additional information or benefits

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

 Difference

-0.72

haw

-0.69

 Dems

-0.68

izens

-0.64

 Grape

-0.62

iste

-0.62

apego

-0.61

 Inher

-0.60

ogun

-0.59

 Pist

-0.59

POSITIVE LOGITS

 thereto

0.71

olkien

0.70

ĸļ

0.66

 receiving

0.64

dating

0.64

ãģĨ

0.62

udes

0.62

 materially

0.62

arez

0.62

 providing

0.62

Activations Density 0.013%

phrases related to the addition or specifying something extra

references to additional information or benefits

No Comments

No Known Activations