INDEX

Explanations

question phrases and expressions probing for explanations or implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

tero

-0.19

ardown

-0.17

roj

-0.15

ython

-0.14

 tagName

-0.14

ÐĤ

-0.14

orris

-0.14

idity

-0.14

rient

-0.14

chner

-0.13

POSITIVE LOGITS

 exactly

0.16

Pax

0.16

 Exactly

0.14

egen

0.14

 YOUR

0.14

ãĥ¼ãĥĲ

0.14

Nou

0.14

iego

0.13

 egret

0.13

ops

0.13

Activations Density 0.428%

question phrases and expressions probing for explanations or implications

No Comments

No Known Activations