INDEX

Explanations

references to specific individuals and names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

.Bunifu

-0.17

omi

-0.15

ept

-0.15

áº¹p

-0.14

 Zhou

-0.14

ferred

-0.14

 craw

-0.13

ccd

-0.13

reau

-0.13

 limit

-0.13

POSITIVE LOGITS

æ°ı

0.17

 Sisters

0.15

_rw

0.15

ario

0.15

å§ĵ

0.15

 ìĶ¨

0.15

å®¶

0.14

 Brothers

0.14

abin

0.14

ìĶ¨

0.14

Activations Density 0.031%

references to specific individuals and names

No Comments

No Known Activations