INDEX

Explanations

elements related to numerical information, specifically distances and measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uzzi

-0.18

 Maduro

-0.17

McB

-0.16

 Mechanics

-0.16

 Meteor

-0.15

 Monetary

-0.15

 Mech

-0.15

Mov

-0.15

MetaData

-0.15

_mono

-0.15

POSITIVE LOGITS

 miles

0.71

 mile

0.71

 Miles

0.68

 Mile

0.64

-mile

0.62

mile

0.57

 mileage

0.55

mi

0.47

Âłmiles

0.44

 milestone

0.43

Activations Density 0.108%

elements related to numerical information, specifically distances and measurements

No Comments

No Known Activations