INDEX

Explanations

references to organizational features and practical aspects of packing

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

abr

-0.14

éĺ¶

-0.14

cio

-0.14

è¹

-0.14

HR

-0.14

 ÑĦÐ¾ÑĢ

-0.14

uru

-0.14

 burner

-0.13

 churches

-0.13

POSITIVE LOGITS

 emptied

0.19

ÑĢÐ°ÑĤÐ¸

0.19

 Empty

0.16

ÑĢÐ°ÑĤÐ½Ð¾

0.16

Storage

0.16

 Storage

0.16

 capacity

0.16

Capacity

0.15

 empty

0.15

ãĤ«ãĥ¼

0.15

Activations Density 0.066%

references to organizational features and practical aspects of packing

No Comments

No Known Activations