INDEX

Explanations

references to championships and winning achievements in sports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

urd

-0.16

amina

-0.16

ensa

-0.16

ValueCollection

-0.15

UNE

-0.15

usk

-0.14

enko

-0.14

alla

-0.14

bero

-0.14

 Bylo

-0.14

POSITIVE LOGITS

osition

0.16

Dos

0.15

Î¼Î¹Ïĥ

0.15

OSC

0.15

ucken

0.15

 Code

0.15

gest

0.15

 interference

0.14

kip

0.14

Activations Density 0.093%

references to championships and winning achievements in sports

No Comments

No Known Activations