INDEX

Explanations

instances of user engagement through comments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

nothrow

-0.15

 agre

-0.15

cliffe

-0.15

opal

-0.15

lect

-0.15

 Bloc

-0.15

ÏĬÎº

-0.15

 kÃ½

-0.14

alleries

-0.14

lec

-0.14

POSITIVE LOGITS

auth

0.15

 IOCTL

0.15

ï¿

0.14

usan

0.14

ple

0.13

ixo

0.13

 PropertyChanged

0.13

bens

0.13

 Authentic

0.13

intro

0.13

Activations Density 0.003%

instances of user engagement through comments

No Comments

No Known Activations