INDEX

Explanations

military-related terms and ranks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

anca

-0.16

steen

-0.15

 reins

-0.15

zdy

-0.15

ladu

-0.15

udo

-0.14

chwitz

-0.14

zsche

-0.14

ynamodb

-0.14

ardo

-0.13

POSITIVE LOGITS

Fur

0.15

abe

0.15

 Granite

0.14

 Sawyer

0.13

è¥

0.13

-equipped

0.13

ouz

0.13

Activations Density 0.033%

military-related terms and ranks

No Comments

No Known Activations