INDEX

Explanations

technical language and terms related to manuals and specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

qus

-0.19

alat

-0.15

ifter

-0.14

 Wright

-0.14

imedia

-0.14

ÑĢÐ°Ð¹

-0.14

 Hein

-0.14

asaki

-0.13

.Media

-0.13

udden

-0.13

POSITIVE LOGITS

ĭ

0.17

åŃĿ

0.14

ç»¼åĲĪ

0.14

hiba

0.14

inha

0.14

ÅĻez

0.14

Ã¡no

0.13

EFR

0.13

è¬

0.13

³

0.13

Activations Density 0.003%

technical language and terms related to manuals and specifications

No Comments

No Known Activations