INDEX

Explanations

code-related comments and metadata in a programming context

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

arl

-0.16

ino

-0.15

issen

-0.15

sep

-0.14

lag

-0.14

ÐµÐ³

-0.14

unn

-0.14

èŃ·

-0.13

]("

-0.13

usic

-0.13

POSITIVE LOGITS

LOT

0.18

adden

0.16

.jquery

0.16

uyo

0.15

anni

0.15

idelberg

0.14

krom

0.14

uforia

0.13

CTR

0.13

âĢı

0.13

Activations Density 0.031%

code-related comments and metadata in a programming context

No Comments

No Known Activations