INDEX

Explanations

references to page numbers and bibliographic elements in texts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

aket

-0.16

INDER

-0.15

UDO

-0.14

udo

-0.14

alue

-0.14

ì

-0.14

dden

-0.14

orm

-0.14

aset

-0.13

POSITIVE LOGITS

.scalablytyped

0.14

 Colleg

0.14

obel

0.14

 Roose

0.14

Da

0.14

bulan

0.13

*[

0.13

ìĨ

0.13

elson

0.13

Å¾en

0.13

Activations Density 0.037%

references to page numbers and bibliographic elements in texts

No Comments

No Known Activations