INDEX

Explanations

technical terms and jargon related to software or technology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ourn

-0.15

ouncer

-0.15

aktu

-0.15

APS

-0.14

snow

-0.14

 Grip

-0.14

at

-0.13

 Snow

-0.13

 lÃ¶

-0.13

avier

-0.13

POSITIVE LOGITS

thora

0.16

suma

0.16

INVAL

0.15

%A

0.14

eref

0.14

Ú¯Ø§ÙĩÛĮ

0.14

ISTA

0.14

 Ð¿Ð¾Ð´Ð²

0.13

_Impl

0.13

Activations Density 0.093%

technical terms and jargon related to software or technology

No Comments

No Known Activations