INDEX

Explanations

names of individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

ìĤ¼

-0.15

Occurs

-0.14

ÙĳÙĩ

-0.13

 GenerationType

-0.13

íĹĮ

-0.12

 olduÄŁuna

-0.12

 sayÄ±lÄ±

-0.12

StackSize

-0.12

İ

-0.12

/sidebar

-0.11

POSITIVE LOGITS

on

0.29

 interviews

0.23

 from

0.23

(@

0.22

’s

0.22

 blogs

0.22

in

0.21

 here

0.21

on

0.21

_the

0.21

Activations Density 0.684%

names of individuals

No Comments

No Known Activations