Latxa 7B V1 By HiTZ: Benchmarks, Features and Detailed Analysis. Insights on Latxa 7B V1.

Arxiv:1910.09700 Arxiv:2112.10668 Arxiv:2308.16884 Autotrain compatible Dataset:hitz/euscrawl En Endpoints compatible Eu Llama Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/HiTZ/latxa-7b-v1

Latxa 7B V1 Benchmarks

LLME Score: 0.17106

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Latxa 7B V1 Parameters and Internals

Model Type

Language model

Use Cases

Areas:

Research, Commercial applications

Additional Notes

Model is outdated; newer versions available in HuggingFace collection.

Supported Languages

en (supported), eu (supported)

Training Details

Data Sources:

Euscrawl v1, 100K documents of English data from the Pile dataset

Data Volume:

1.72M documents, 288M words, 2.1GiB of uncompressed text

Methodology:

Trained using GPT-Neox library on HPC CINECA computing cluster

Context Length:

4096

Training Time:

1680 steps, effective batch size of 2M tokens, 6475.52 GPU hours

Hardware Used:

4x A100 64Gb nodes

LLM Name	Latxa 7B V1
Repository 🤗	https://huggingface.co/HiTZ/latxa-7b-v1
Model Size	7b
Required VRAM	13.5 GB
Updated	2024-12-22
Maintainer	HiTZ
Model Type	llama
Model Files	10.0 GB: 1-of-2 3.5 GB: 2-of-2
Supported Languages	eu en
Model Architecture	LlamaForCausalLM
License	llama2
Context Length	4096
Model Max Length	4096
Transformers Version	4.31.0
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32000
Torch Data Type	bfloat16

Quantized Models of the Latxa 7B V1

Model	Likes	Downloads	VRAM
Latxa 7B Instruct	0	5	13 GB

Best Alternatives to Latxa 7B V1

Best Alternatives	Context / RAM	Downloads	Likes
...1M 1000000ctx AEZAKMI 3 1 1702	1024K / 13.5 GB	85	1
... Qwen2.5llamaify 7B V23.1 200K	195K / 15.2 GB	2935	0
SuperNeuralDreadDevil 8B	128K / 16.1 GB	155	1
Yarn Llama 2 7B 128K	128K / 13.5 GB	2262	40
LLaMA 7B PoSE YaRN 128K	128K / 13.5 GB	11	3
LLaMA 7B PoSE Linear 96K	96K / 27 GB	10	2
LLaMA 7B PoSE YaRN 96K	96K / 13.5 GB	11	1
Chat Llama2 7B 80K	80K / 13.8 GB	26	0
Llama2 7B 80K	80K / 13.8 GB	25	0
Lloma Step400	64K / 13.5 GB	80	0

Note: green Score (e.g. "73.2") means that the model is better than HiTZ/latxa-7b-v1.

Rank the Latxa 7B V1 Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

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

What open-source LLMs or SLMs are you in search of? 40123 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, OpenCompass and various public git repos.

Release v20241217

Support LLM Explorer