What are the hardware requirements for Llama 3 Instruction Constructionsafety Layertuning?

Llama 3 Instruction Constructionsafety Layertuning requires approximately 16.1 GB of VRAM and supports a context window of 8K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

How does Llama 3 Instruction Constructionsafety Layertuning perform on standard benchmarks?

Llama 3 Instruction Constructionsafety Layertuning has the following published scores: MMLU 55.67. Compare against reference models on this page or on the LLM Explorer leaderboards.

Llama 3 Instruction Constructionsafety Layertuning by DBCMLAB — VRAM 16.1GB, 8K context

Name: Llama 3 Instruction Constructionsafety Layertuning
Author: DBCMLAB

Llama 3 Instruction Constructionsafety Layertuning is an open-source language model by DBCMLAB. Features: 8b LLM, VRAM: 16.1GB, Context: 8K, License: cc-by-nc-4.0, Instruction-Based, LLM Explorer Score: 0.15, Arc: 55.5, HellaSwag: 75.1, MMLU: 55.7, GSM8K: 33.6.

Conversational Deploy:azure Endpoints compatible Facebook Instruct Ko Llama Llama3 Meta Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/DBCMLAB/Llama-3-instruction-constructionsafety-layertuning

Llama 3 Instruction Constructionsafety Layertuning Parameters and Internals

Model Type

fine-tuned

Use Cases

Applications:

Construction safety training

Primary Use Cases:

Providing safety guidelines in construction

Limitations:

Possibly limited to construction safety context

Supported Languages

Korean (high)

Training Details

Data Sources:

QA datasets, RAW data from Korea Occupational Safety and Health Agency (KOSHA)

Data Volume:

11,000 examples

Methodology:

Full parameter tuning; layers 0, 30, and 31 replaced post-training with base model parameters

Hardware Used:

2xA100GPU (80GB)

LLM Name	Llama 3 Instruction Constructionsafety Layertuning
Repository 🤗	https://huggingface.co/DBCMLAB/Llama-3-instruction-constructionsafety-layertuning
Model Size	8b
Required VRAM	16.1 GB
Updated	2026-08-24
Maintainer	DBCMLAB
Model Type	llama
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-4 5.0 GB: 2-of-4 4.9 GB: 3-of-4 1.2 GB: 4-of-4
Supported Languages	ko
Model Architecture	LlamaForCausalLM
License	cc-by-nc-4.0
Context Length	8192
Model Max Length	8192
Transformers Version	4.41.0
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<\|end_of_text\|>
Vocabulary Size	128256
Torch Data Type	bfloat16

Rank the Llama 3 Instruction Constructionsafety Layertuning capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Llama 3 Instruction Constructionsafety Layertuning

Best Alternatives	Context / RAM	Downloads	Likes
...otron 8B UltraLong 4M Instruct	4192K / 32.1 GB	1135	125
UltraLong Thinking	4192K / 16.1 GB	2	3
...a 3.1 8B UltraLong 4M Instruct	4192K / 32.1 GB	176	24
...a 3.1 8B UltraLong 2M Instruct	2096K / 32.1 GB	875	9
...otron 8B UltraLong 2M Instruct	2096K / 32.1 GB	124	18
Cthulhu 8B V1.4	1048K / 16.1 GB	10	10
...raLong 1M Instruct Abliterated	1048K / 32.1 GB	4	9
...a 3.1 8B UltraLong 1M Instruct	1048K / 32.1 GB	1387	29
...otron 8B UltraLong 1M Instruct	1048K / 32.1 GB	702	59
Zero Llama 3.1 8B Beta6	1048K / 16.1 GB	7	1

Note: green Score (e.g. "73.2") means that the model is better than DBCMLAB/Llama-3-instruction-constructionsafety-layertuning.