INDEX

Explanations

references to pop culture icons and phenomena

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

 Foley

-0.15

Hu

-0.14

hu

-0.14

Ð¾Ð»Ð¾Ñģ

-0.14

ishi

-0.14

 Elvis

-0.13

xic

-0.13

ele

-0.13

ugi

-0.13

rant

-0.13

POSITIVE LOGITS

irut

0.18

irk

0.15

/terms

0.15

iral

0.15

ilog

0.15

iah

0.15

ipeg

0.14

hone

0.14

-Cal

0.14

apter

0.14

Activations Density 0.463%

references to pop culture icons and phenomena

No Comments

No Known Activations