Avg. Total Time
14.16s
Avg. TTFT
11.36s
Avg. Prefill TPS
117.49
Avg. Gen TPS
22.18
Context Size
32768
Quantization
r64
Engine
aphrodite
Creation Method
Merge
Model Type
Llama70B
Chat Template
Llama 3
Reasoning
No
Vision
No
Parameters
70B
Added At
12/22/2024
base_model: [] library_name: transformers tags:
This is a merge of pre-trained language models created using mergekit.
This model was merged using the SLERP merge method.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
base_model: ../mergekit-output/new-dawn-ultra-llama3-70b-32K-v1.0.yaml/new-dawn-ultra-llama3-70b-v0.16-32K
dtype: float32
merge_method: slerp
parameters:
t:
- value: 0.5
slices:
- sources:
- layer_range: [0, 80]
model: ../mergekit-output/new-dawn-ultra-llama3-70b-32K-v1.0.yaml/new-dawn-ultra-llama3-70b-v0.16-32K
- layer_range: [0, 80]
model: ../mergekit-output/new-dawn-ultra-llama3-70b-32K-v1.0.yaml/new-dawn-ultra-llama3-70b-v0.18-32K