INDEX

Explanations

language related to cost reduction and efficiency improvement in various contexts

oai_token-act-pair · gpt-3.5-turbo

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ylum

-0.85

 belonged

-0.81

 belong

-0.76

 belongs

-0.71

 solemn

-0.70

NAME

-0.69

 Assange

-0.68

encia

-0.67

arag

-0.66

Names

-0.64

POSITIVE LOGITS

 decrease

1.48

Decre

1.43

 decreases

1.40

 reduction

1.37

 reductions

1.36

 reducing

1.34

 lowering

1.34

 increase

1.30

Increase

1.29

 increases

1.28

Activations Density 0.728%

language related to cost reduction and efficiency improvement in various contexts

No Comments

No Known Activations