INDEX

Explanations

Java package and class names from a software project

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ekk

-0.14

Ð·Ð°Ð±

-0.14

-Bold

-0.13

ehler

-0.13

efon

-0.13

 Mage

-0.13

itm

-0.12

 Cher

-0.12

.qml

-0.12

Ras

-0.12

POSITIVE LOGITS

.util

0.18

utils

0.16

iform

0.16

;↵↵

0.15

.model

0.15

kee

0.15

ummer

0.15

.common

0.15

jure

0.15

framework

0.14

Activations Density 0.091%

Java package and class names from a software project

No Comments

No Known Activations