INDEX

Explanations

references to downloadable materials and resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

rana

-0.15

ippy

-0.15

itesse

-0.15

åŃĹå¹ķ

-0.15

atura

-0.14

idea

-0.14

 amalg

-0.14

chemy

-0.14

cola

-0.14

 Scalia

-0.14

POSITIVE LOGITS

attern

0.16

Ø¨ÙĬ

0.15

 Fund

0.14

addy

0.14

>Returns

0.14

.OnClickListener

0.14

aska

0.14

ADO

0.14

 handbook

0.13

uada

0.13

Activations Density 0.029%

references to downloadable materials and resources

No Comments

No Known Activations