INDEX

Explanations

references to military conflicts and the involved parties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 English

-0.17

 phenomena

-0.16

 english

-0.15

-0.14

US

-0.14

 McCabe

-0.14

Hij

-0.14

oso

-0.14

hc

-0.13

 Join

-0.13

POSITIVE LOGITS

 presence

0.27

 forces

0.23

 Presence

0.22

 occupation

0.21

presence

0.21

-controlled

0.20

 advance

0.20

Presence

0.20

-held

0.19

-inst

0.19

Activations Density 0.110%

references to military conflicts and the involved parties

No Comments

No Known Activations