INDEX

Explanations

image file references and URLs

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

ryn

-0.17

atted

-0.14

 Ð±ÐµÑĢ

-0.14

tplib

-0.14

gra

-0.14

Ð±ÐµÑĢ

-0.14

iram

-0.14

allax

-0.13

.Guna

-0.13

 atIndex

-0.13

POSITIVE LOGITS

ASC

0.15

ENTA

0.15

finity

0.15

thouse

0.14

enta

0.14

Fal

0.14

 temps

0.14

å¤§ä¼ļ

0.14

Î±Ïĥ

0.14

CEF

0.13

Activations Density 0.002%

image file references and URLs

No Comments

No Known Activations