INDEX

Explanations

fragments of credit or attribution in text

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

Overview

-0.15

 Appeals

-0.14

 Overview

-0.14

orean

-0.13

zer

-0.13

ven

-0.13

 Brass

-0.13

Viewport

-0.13

 overview

-0.13

jang

-0.13

POSITIVE LOGITS

agina

0.16

 dÃ¼

0.14

apus

0.14

Î±Î¹Î¿

0.13

atÃ³

0.13

licer

0.13

odom

0.13

 GetEnumerator

0.13

Gatt

0.13

adays

0.13

Activations Density 0.009%

fragments of credit or attribution in text

No Comments

No Known Activations