INDEX

Explanations

phrases related to the extent or boundaries of something

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "scope."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

weet

-0.79

ãĤ°

-0.70

edient

-0.67

--------------

-0.66

elin

-0.66

washed

-0.65

itates

-0.65

itating

-0.65

haps

-0.64

 Valentine

-0.64

POSITIVE LOGITS

 scope

3.99

 Scope

2.38

scope

2.31

Scope

2.21

 breadth

1.52

 scale

1.42

 extent

1.22

 severity

1.02

 context

1.02

 bounds

1.02

Activations Density 0.013%

phrases related to the extent or boundaries of something

references to the concept of "scope."

No Comments

No Known Activations