INDEX

Explanations

mentions of the word "rep" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

ering

-0.17

erland

-0.15

ãģ¨ãģĵãĤį

-0.15

nut

-0.15

lette

-0.15

 analog

-0.15

amaz

-0.14

ered

-0.14

anut

-0.14

pack

-0.14

POSITIVE LOGITS

tember

0.19

Ãºblica

0.17

UGE

0.15

oster

0.15

OLVE

0.15

orent

0.14

oxid

0.14

aldo

0.14

owned

0.14

Ð¾Ð²Ð¸Ñĩ

0.14

Activations Density 0.033%

mentions of the word "rep" and its variations

No Comments

No Known Activations