INDEX

Explanations

references to specific projects

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

clus

-0.16

unker

-0.16

arking

-0.15

 ØŃØ±ÙģÙĩ

-0.14

áº¤

-0.14

aravel

-0.14

uong

-0.14

lej

-0.14

.edu

-0.14

allas

-0.14

POSITIVE LOGITS

 Wikimedia

0.17

relative

0.15

Pru

0.15

 Huck

0.14

tpl

0.14

Zub

0.14

yu

0.14

 Esper

0.14

-runtime

0.13

.si

0.13

Activations Density 0.005%

references to specific projects

No Comments

No Known Activations