INDEX

Explanations

references to historical buildings and their transformations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

strup

-0.17

-0.16

fetchAll

-0.14

Fu

-0.14

 Exped

-0.14

EDA

-0.14

ØµÙģ

-0.14

 sÃłng

-0.14

Ø±Ø§Ùģ

-0.14

fetch

-0.14

POSITIVE LOGITS

former

0.15

Ø§Ø¶

0.15

Ø§Ø¶ÙĬ

0.15

 former

0.14

 Former

0.14

enler

0.14

jev

0.13

arro

0.13

chrift

0.13

 overd

0.13

Activations Density 0.070%

references to historical buildings and their transformations

No Comments

No Known Activations