INDEX

Explanations

references to specific items, their organization, or grouping

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 GenerationType

-0.15

 Both

-0.14

aza

-0.13

 competing

-0.13

 precursor

-0.13

reed

-0.13

gz

-0.13

ella

-0.13

inned

-0.13

èĪ

-0.13

POSITIVE LOGITS

 ones

0.29

one

0.26

åħ¶ä¸Ń

0.26

ä¹ĭä¸Ģ

0.24

çļĦä¸Ģä¸ª

0.23

 eines

0.21

 ÛĮÚ©ÛĮ

0.21

 fourth

0.20

ç¬¬

0.20

 oldest

0.20

Activations Density 0.442%

references to specific items, their organization, or grouping

No Comments

No Known Activations