INDEX

Explanations

references to military operations and personnel

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

 army

-0.19

 terrain

-0.18

 Terrain

-0.17

terrain

-0.16

 Border

-0.16

pup

-0.15

 gran

-0.15

 luggage

-0.15

 armies

-0.15

ibri

-0.15

POSITIVE LOGITS

èī¦

0.23

èĪ°

0.21

USS

0.20

 Naval

0.20

 Navy

0.20

 underway

0.20

 Fleet

0.19

Zum

0.18

 deck

0.17

 Ø§ÙĦØ¨ØŃØ±

0.17

Activations Density 0.096%

references to military operations and personnel

No Comments

No Known Activations