INDEX

Explanations

references to magazine issues and publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

/Dk

-0.20

-0.17

izza

-0.15

ogl

-0.15

MLS

-0.14

ovel

-0.14

 benches

-0.14

haus

-0.14

ashes

-0.14

idden

-0.14

POSITIVE LOGITS

anas

0.17

ibase

0.15

Ã¡st

0.15

ãĤ¯ãĥĪ

0.15

.addAll

0.14

blade

0.14

Å¡t

0.14

ãģªãģıãģª

0.14

-available

0.14

alse

0.14

Activations Density 0.019%

references to magazine issues and publications

No Comments

No Known Activations