INDEX

Explanations

references to scholarly work or citations

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

///<

-0.17

ä

-0.15

.communic

-0.15

adil

-0.15

unan

-0.14

eea

-0.14

pret

-0.14

oster

-0.14

isi

-0.14

undef

-0.14

POSITIVE LOGITS

 Paid

0.18

 Fluent

0.16

AVL

0.15

ANTE

0.15

 Lakes

0.15

åĨĮ

0.15

chos

0.15

xea

0.15

 Sweat

0.14

 Await

0.14

Activations Density 0.067%

references to scholarly work or citations

No Comments

No Known Activations