INDEX

Explanations

terms related to organizational structures and programs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ava

-0.16

cing

-0.16

ody

-0.14

orton

-0.14

ÑĢÐ¾Ñĩ

-0.14

 hers

-0.14

ruise

-0.14

uchs

-0.14

ours

-0.13

iox

-0.13

POSITIVE LOGITS

herits

0.17

ManagerInterface

0.16

iveau

0.16

Ã©v

0.16

Ã¼lÃ¼

0.15

åĿĹ

0.14

rador

0.14

IPLE

0.14

arLayout

0.14

tiler

0.14

Activations Density 0.131%

terms related to organizational structures and programs

No Comments

No Known Activations