INDEX

Explanations

references to resources or documentation links

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 DISPATCH

-0.16

HEME

-0.15

 iddi

-0.15

Ø§ÛĮØ¹

-0.15

Ã¡tis

-0.15

ãĥ§

-0.14

radient

-0.14

ï¼ļ"

-0.14

oth

-0.13

PathComponent

-0.13

POSITIVE LOGITS

 Golf

0.14

åİŁ

0.14

akers

0.14

inic

0.14

dbl

0.14

 Kind

0.14

aler

0.14

 Ð³ÑĢÐ°

0.14

 KIND

0.13

 Alloy

0.13

Activations Density 0.015%

references to resources or documentation links

No Comments

No Known Activations