INDEX

Explanations

elements related to cultural or geographical names

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

uer

-0.17

 Watts

-0.16

atem

-0.15

è¦

-0.15

zer

-0.15

à¸¹à¸£

-0.14

Å¡tÄĽ

-0.14

ulle

-0.14

mpar

-0.14

POSITIVE LOGITS

jet

0.17

PELL

0.16

ãĥ³ãĤº

0.15

jes

0.15

kovi

0.15

ista

0.15

ij

0.15

Äĩ

0.15

 pomoc

0.14

.LookAndFeel

0.14

Activations Density 0.013%

elements related to cultural or geographical names

No Comments

No Known Activations