INDEX

Explanations

unique and specific identifiers, potentially for regions or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

Ó

-0.15

arshal

-0.15

/=

-0.14

otes

-0.14

ditor

-0.14

Ò

-0.14

.Rendering

-0.14

kova

-0.14

actal

-0.13

ÐĲÑĢÑħÑĸÐ²

-0.13

POSITIVE LOGITS

 menus

0.21

 MENU

0.20

 menu

0.19

Menu

0.19

 Menu

0.18

/menu

0.17

.menu

0.16

 Lump

0.16

-menu

0.16

_menus

0.15

Activations Density 0.000%

unique and specific identifiers, potentially for regions or entities

No Comments

No Known Activations