INDEX

Explanations

titles and roles related to academic and research positions

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

.gov

-0.16

gov

-0.15

akis

-0.15

.MixedReality

-0.14

otionEvent

-0.14

asher

-0.14

æĽ¸é¤¨

-0.14

umba

-0.14

ade

-0.14

POSITIVE LOGITS

 Frontier

0.26

 Graduate

0.23

 Hose

0.23

Grad

0.21

 graduate

0.20

 frontier

0.20

 Grad

0.19

 Okay

0.18

 Sophia

0.17

 Koch

0.17

Activations Density 0.019%

titles and roles related to academic and research positions

No Comments

No Known Activations