INDEX

Explanations

technical terms and identifiers related to data structures and software functionality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÑĸÑģ

-0.16

reek

-0.16

à¸¨à¸²à¸ªà¸ķà¸£

-0.15

iÅ¡

-0.15

avana

-0.15

sik

-0.15

ensex

-0.15

reo

-0.14

isoft

-0.14

vala

-0.14

POSITIVE LOGITS

paz

0.15

 Daniels

0.14

Yaz

0.14

Mil

0.13

ar

0.13

ulous

0.13

ifr

0.13

 æĢ

0.13

 Tanner

0.13

''

0.13

Activations Density 0.576%

technical terms and identifiers related to data structures and software functionality

No Comments

No Known Activations