INDEX

Explanations

expressions emphasizing frequency or specific instances

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

erken

-0.16

ä¸įäºĨ

-0.15

å¤§äºº

-0.14

/autoload

-0.14

amura

-0.14

jos

-0.14

/lists

-0.13

ControlEvents

-0.13

Ãºng

-0.13

ÐµÑĢÐ°Ð¿

-0.13

POSITIVE LOGITS

allel

0.17

 inch

0.15

 iota

0.15

airo

0.14

athon

0.14

 Cairo

0.14

redi

0.14

 Bans

0.14

uard

0.14

iger

0.13

Activations Density 0.052%

expressions emphasizing frequency or specific instances

No Comments

No Known Activations