INDEX

Explanations

references to specific years or time periods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 lately

-0.19

 recently

-0.19

 tonight

-0.18

 recent

-0.17

 Recently

-0.17

setContent

-0.16

 tomorrow

-0.16

Recently

-0.16

 latter

-0.16

 earlier

-0.16

POSITIVE LOGITS

 ÙħÛĮÙĦØ§Ø¯ÛĮ

0.23

ago

0.18

 when

0.16

INO

0.15

atories

0.15

ajo

0.15

/right

0.15

eds

0.15

-end

0.15

-present

0.14

Activations Density 0.042%

references to specific years or time periods

No Comments

No Known Activations