INDEX

Explanations

code-related concepts and annotations in Java

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

kl

-0.16

 Pols

-0.14

Nap

-0.14

Je

-0.14

lycer

-0.14

-0.13

.NET

-0.13

den

-0.13

borg

-0.13

POSITIVE LOGITS

 logger

0.19

.logger

0.18

elper

0.17

ocket

0.17

ante

0.16

 Logger

0.15

(logger

0.14

mana

0.14

undry

0.14

 Faces

0.14

Activations Density 0.473%

code-related concepts and annotations in Java

No Comments

No Known Activations