INDEX

Explanations

locations described in terms of distance, such as in kilometers or miles away from certain places

oai_token-act-pair · gpt-3.5-turbo

numerical values, particularly distances or measurements

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

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

 congr

-0.71

 Proc

-0.67

 disadvant

-0.66

 proc

-0.66

 presumption

-0.63

 vanity

-0.61

 ransomware

-0.61

 traged

-0.61

 Myster

-0.60

 facult

-0.60

POSITIVE LOGITS

 kilometres

1.80

 miles

1.80

 kilometers

1.79

km

1.67

km

1.65

 kilomet

1.37

 mile

1.32

 hectares

1.30

mile

1.27

 acres

1.25

Activations Density 0.105%

locations described in terms of distance, such as in kilometers or miles away from certain places

numerical values, particularly distances or measurements

No Comments

No Known Activations