INDEX

Explanations

comparative phrases indicating differences between subjects or entities

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

aos

-0.18

}elseif

-0.15

åĬ¡

-0.15

iazza

-0.15

ipel

-0.14

iaz

-0.14

cke

-0.14

mma

-0.13

¡

-0.13

iah

-0.13

POSITIVE LOGITS

rod

0.16

Rod

0.15

ingleton

0.15

rod

0.15

 other

0.15

MATRIX

0.15

other

0.15

 sarc

0.15

arent

0.14

 others

0.14

Activations Density 0.163%

comparative phrases indicating differences between subjects or entities

No Comments

No Known Activations