INDEX

Explanations

punctuated phrases indicating preferences or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

Dah

-0.18

prung

-0.15

 trap

-0.15

rade

-0.14

ANGLE

-0.14

 dane

-0.14

lov

-0.14

 Blind

-0.14

ditor

-0.14

AlmostEqual

-0.14

POSITIVE LOGITS

 thÃ¬

0.21

then

0.19

 then

0.19

çļĦè¯Ŀ

0.19

 chances

0.18

 entonces

0.18

 entÃ£o

0.17

 maka

0.17

 alors

0.16

éĤ£ä¹Ī

0.16

Activations Density 0.160%

punctuated phrases indicating preferences or conditions

No Comments

No Known Activations