INDEX

Explanations

specific file types or formats often related to documents and data management

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

LOCKS

-0.15

essian

-0.15

mmc

-0.14

glm

-0.14

allas

-0.14

abe

-0.13

Ã§o

-0.13

iÃ§

-0.13

lsru

-0.13

aje

-0.13

POSITIVE LOGITS

itud

0.15

byte

0.14

odore

0.14

hiba

0.14

oningen

0.14

rega

0.13

IFY

0.13

 Neville

0.13

á»©a

0.13

ronic

0.12

Activations Density 0.194%

specific file types or formats often related to documents and data management

No Comments

No Known Activations