INDEX

Explanations

numerical values and terms related to military operations and logistical details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 voksne

-0.16

 WWII

-0.16

ocaust

-0.14

ãĥ³ãĥĶ

-0.13

Ã£i

-0.13

 pisc

-0.13

ãĥ³ãĤ¯

-0.13

 Missile

-0.13

-Assad

-0.13

à¸±à¸ģà¸Ķ

-0.13

POSITIVE LOGITS

 General

0.29

 news

0.26

 reinforcements

0.24

 word

0.23

 Major

0.23

 Lord

0.23

Gen

0.21

 Admiral

0.21

 supplies

0.21

General

0.21

Activations Density 0.098%

numerical values and terms related to military operations and logistical details

No Comments

No Known Activations