INDEX

Explanations

questions that begin with the word "who."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

mente

-0.23

ted

-0.22

tings

-0.17

uen

-0.17

uros

-0.16

ting

-0.16

tti

-0.16

halb

-0.15

EATURE

-0.15

chen

-0.15

POSITIVE LOGITS

 else

0.22

soever

0.17

oping

0.17

áº¡t

0.16

osh

0.16

inspace

0.16

_else

0.15

-Ð½Ð¸Ð±ÑĥÐ´ÑĮ

0.15

apk

0.15

EVER

0.15

Activations Density 0.037%

questions that begin with the word "who."

No Comments

No Known Activations