INDEX

Explanations

phrases expressing endorsement or assistance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

UEL

-0.17

ALA

-0.16

 anchor

-0.15

 Ð³ÑĢÐ¾Ð¼

-0.15

ï»¿#

-0.14

 ìı

-0.13

 sona

-0.13

kud

-0.13

inson

-0.13

ÑĢÐ°Ñĩ

-0.13

POSITIVE LOGITS

iaux

0.22

of

0.15

çĳŁ

0.14

 kick

0.14

Sector

0.14

.virtual

0.14

 Sector

0.14

ansen

0.13

atoria

0.13

 interf

0.13

Activations Density 0.031%

phrases expressing endorsement or assistance

No Comments

No Known Activations