INDEX

Explanations

references to conflict and violence, particularly related to terrorism and illegal settlements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

 Mercy

-0.16

adm

-0.15

Anchor

-0.15

aram

-0.15

ovnÃŃ

-0.15

oples

-0.14

 kostenlose

-0.14

 Kemal

-0.13

legg

-0.13

Dog

-0.13

POSITIVE LOGITS

endra

0.20

Å¼

0.15

estroy

0.14

Lub

0.14

 Chore

0.14

/ng

0.13

ë°ĶìĿ´

0.13

adena

0.13

riot

0.13

å¯¸

0.13

Activations Density 0.006%

references to conflict and violence, particularly related to terrorism and illegal settlements

No Comments

No Known Activations