INDEX

Explanations

descriptions of community features and amenities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 buildings

-0.17

 Ð·Ð´

-0.17

ello

-0.15

éŀ

-0.15

 house

-0.15

zin

-0.15

 shed

-0.14

thon

-0.14

 Buildings

-0.14

 premises

-0.14

POSITIVE LOGITS

 meeting

0.22

 breakout

0.20

 Areas

0.19

meeting

0.19

 Meeting

0.19

 commons

0.18

 writable

0.18

 areas

0.18

 quiet

0.17

 conference

0.17

Activations Density 0.070%

descriptions of community features and amenities

No Comments

No Known Activations