Etheria 55B V0.1 By Steelskull: Benchmarks, Features and Detailed Analysis. Insights on Etheria 55B V0.1.

Merged Model Arxiv:2306.01708 Arxiv:2311.03099 Autotrain compatible Endpoints compatible Etheria Llama Model-index Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Steelskull/Etheria-55b-v0.1

Etheria 55B V0.1 Benchmarks

ARC: 65.1 vs 96.7 (so35)^-32.7%

HellaSwag: 81.93 vs 95.3 (gpt4)^-14%

MMLU: 73.66 vs 88.3 (so35)^-16.6%

TruthfulQA: 56.16 vs 59 (gpt4)^-4.8%

WinoGrande: 76.09 vs 87.5 (gpt4)^-13%

GSM8K: 35.18 vs 96.4 (so35)^-63.5%

LLME Score: 0.17803

^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)

Etheria 55B V0.1 (Steelskull/Etheria-55b-v0.1)

Etheria 55B V0.1 Parameters and Internals

Model Type

text-generation

Additional Notes

An attempt to make a functional goliath style merge to create Etheria 55b-200k with two yi-34b-200k models. Due to the merge, it 'theoretically' should have a context of 200k, but starting at 32k is recommended.

Input Output

Input Format:

ChatML & Alpaca

LLM Name	Etheria 55B V0.1
Repository 🤗	https://huggingface.co/Steelskull/Etheria-55b-v0.1
Merged Model	Yes
Model Size	55b
Required VRAM	111.2 GB
Updated	2024-10-16
Maintainer	Steelskull
Model Type	llama
Model Files	9.8 GB: 1-of-12 9.8 GB: 2-of-12 9.8 GB: 3-of-12 9.9 GB: 4-of-12 10.0 GB: 5-of-12 9.9 GB: 6-of-12 9.8 GB: 7-of-12 9.8 GB: 8-of-12 9.9 GB: 9-of-12 9.9 GB: 10-of-12 9.9 GB: 11-of-12 2.7 GB: 12-of-12
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	200000
Model Max Length	200000
Transformers Version	4.37.1
Tokenizer Class	LlamaTokenizer
Padding Token	<unk>
Vocabulary Size	64002
Torch Data Type	bfloat16

Quantized Models of the Etheria 55B V0.1

Model	Likes	Downloads	VRAM
Etheria 55B V0.1 GGUF	8	255	20 GB
Etheria 55B V0.1 GPTQ	4	26	29 GB
Etheria 55B V0.1 AWQ	1	14	30 GB

Best Alternatives to Etheria 55B V0.1

Best Alternatives	Context / RAM	Downloads	Likes
Etheria 55B V0.1	195K / 111.2 GB	20	10
SG Raccoon Yi 55B 200K 2.0	195K / 111.3 GB	32	6
SG Raccoon Yi 55B 200K	195K / 111.4 GB	53	4
SG Raccoon Yi 55B	4K / 111.2 GB	56	6
...theria 55B V0.1 3.0bpw H6 EXL2	195K / 21.9 GB	4	2
...theria 55B V0.1 3.5bpw H6 EXL2	195K / 25.3 GB	4	1
Etheria 55B V0.1 GPTQ	195K / 29.2 GB	26	4
Etheria 55B V0.1 AWQ	195K / 30.2 GB	14	1
SG Raccoon Yi 55B GPTQ	4K / 29.2 GB	28	1
SG Raccoon Yi 55B AWQ	4K / 30.2 GB	11	2

Rank the Etheria 55B V0.1 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? 43508 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 v20241227

Support LLM Explorer