INDEX

Explanations

references to educational courses and their placement criteria

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

erson

-0.16

positor

-0.14

 medicine

-0.14

mbH

-0.14

haul

-0.14

íĮĶ

-0.14

agenda

-0.14

erosis

-0.14

ÐµÑģÐ¸

-0.14

ANJI

-0.14

POSITIVE LOGITS

 intermediate

0.17

ushi

0.16

iao

0.16

 Introduction

0.16

 Intermediate

0.16

 intro

0.15

 senior

0.15

 Senior

0.15

lish

0.15

 Directed

0.15

Activations Density 0.010%

references to educational courses and their placement criteria

No Comments

No Known Activations