INDEX

Explanations

proper nouns, specifically names and titles associated with individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

ollectors

-0.17

recated

-0.17

evenodd

-0.17

ayÄ±f

-0.17

Ä±lÄ±ÄŁÄ±yla

-0.16

kÃ¢n

-0.16

esModule

-0.16

_mC

-0.16

@nate

-0.16

_mB

-0.15

POSITIVE LOGITS

Con

0.20

 January

0.18

 April

0.18

“

0.18

 March

0.18

 September

0.18

 Czech

0.18

 August

0.18

 English

0.17

May

0.17

Activations Density 0.103%

proper nouns, specifically names and titles associated with individuals

No Comments

No Known Activations