INDEX

Explanations

references to academic articles, including publication details like volume and issue numbers

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

rame

-0.15

 Garten

-0.14

hpp

-0.14

.getD

-0.14

 cmdline

-0.13

 Course

-0.13

 Lips

-0.13

Ð¿Ð¾Ñģ

-0.13

amins

-0.13

afe

-0.13

POSITIVE LOGITS

eka

0.19

 elev

0.15

licht

0.15

bis

0.14

Î¯Ïĥ

0.14

ÐµÐºÐ°ÑĢ

0.14

Łģ

0.14

CAF

0.14

CEO

0.13

Activations Density 0.007%

references to academic articles, including publication details like volume and issue numbers

No Comments

No Known Activations