INDEX

Explanations

expressions indicating strong emotional responses or sentiments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 ÑĥÑĩÐ°ÑģÑĤÐ¸

-0.14

regor

-0.14

 ÄĳÃ¡ng

-0.14

 Äĳá»ĭa

-0.13

 Analyst

-0.13

stalk

-0.13

Ø§Ø¦Ø²

-0.13

ë´ī

-0.13

Ø¨Ø·

-0.13

implified

-0.13

POSITIVE LOGITS

 ageing

0.18

 older

0.17

 aging

0.17

 Elder

0.16

 preserving

0.15

 elder

0.15

Old

0.15

 maint

0.15

 lineage

0.15

 vener

0.15

Activations Density 0.022%

expressions indicating strong emotional responses or sentiments

No Comments

No Known Activations