INDEX

Explanations

references to attacks and assaults, particularly in a political or physical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

rick

-0.18

ãģªãĤī

-0.17

ric

-0.17

autoload

-0.16

estroy

-0.16

ä¸Ī

-0.15

 olsun

-0.15

Ð°Ð½ÑģÐ¸

-0.15

Ð´ÐµÐ¹ÑģÑĤÐ²

-0.14

 dÄ±ÅŁÄ±

-0.14

POSITIVE LOGITS

ively

0.20

iveness

0.17

ive

0.16

ainment

0.16

able

0.15

/mock

0.15

ingly

0.14

avier

0.14

Activations Density 0.053%

references to attacks and assaults, particularly in a political or physical context

No Comments

No Known Activations