INDEX

Explanations

references to geographic locations and community-related topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ibus

-0.16

ostel

-0.15

 Ø¬Ø§Ùħ

-0.15

.mdl

-0.15

 nave

-0.14

ceph

-0.14

 playable

-0.14

.scalablytyped

-0.14

.djang

-0.13

 Ø§ÙĦÙħØ´

-0.13

POSITIVE LOGITS

 Westminster

0.24

 Thornton

0.22

 Wheat

0.22

 Weld

0.21

 Commerce

0.20

GREE

0.20

LONG

0.20

 Parker

0.20

Lov

0.19

 Castle

0.19

Activations Density 0.029%

references to geographic locations and community-related topics

No Comments

No Known Activations