INDEX

Explanations

patterns and structures in code, particularly related to functions and conditional expressions

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

abal

-0.16

.precision

-0.15

/Instruction

-0.15

 ç²¾

-0.15

æ®µ

-0.14

itz

-0.14

ENTIC

-0.14

ãĤ¸ãĤ¢

-0.14

 cáº£nh

-0.14

ibt

-0.14

POSITIVE LOGITS

isphere

0.17

egra

0.16

 quot

0.15

solete

0.15

Playable

0.14

 eskort

0.14

XS

0.14

åĬ¡

0.14

NSS

0.13

eldorf

0.13

Activations Density 0.075%

patterns and structures in code, particularly related to functions and conditional expressions

No Comments

No Known Activations