INDEX

Explanations

years mentioned in a chronological context

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

 January

-0.22

 Winter

-0.21

January

-0.20

 winter

-0.20

 February

-0.19

JAN

-0.19

 March

-0.18

Jan

-0.18

 Christmas

-0.18

Winter

-0.17

POSITIVE LOGITS

itler

0.17

echan

0.16

.hardware

0.15

estival

0.15

 Potion

0.14

ilter

0.14

ëħĦ

0.14

BUG

0.14

]=>

0.13

ãĥ³ãĥĹ

0.13

Activations Density 0.103%

years mentioned in a chronological context

No Comments

No Known Activations