INDEX

Explanations

terms related to scientific and mathematical concepts, particularly in discussions of analytical methods and calculations

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

Ngh

-0.16

abeth

-0.15

culus

-0.14

osate

-0.14

ifact

-0.14

ke

-0.14

ryn

-0.14

 strSQL

-0.14

Ð¾Ð½ÑĮ

-0.13

radu

-0.13

POSITIVE LOGITS

EIF

0.15

åĨĬ

0.14

LBL

0.14

ÂĢÂ

0.14

atory

0.13

Ð½Ð¾ÑģÑĤ

0.13

 hostage

0.13

anos

0.13

iki

0.13

 Serialized

0.13

Activations Density 0.131%

terms related to scientific and mathematical concepts, particularly in discussions of analytical methods and calculations

No Comments

No Known Activations