The following YAML configuration was used to produce this model:
merge_method:linear# use linear so we can include multiple models, albeit at a zero weightparameters:weight:1.0# weight everything as 1 unless specified otherwise - linear with one model weighted at 1 is a no-op like passthroughslices:-sources:-model:cognitivecomputations/dolphin-2.2-70b# embed_tokens comes along with the ride with whatever is the first layerlayer_range: [0, 1]
-model:NousResearch/Nous-Hermes-2-Llama-2-70B# add dummy second model with 0 weight so tokenizer-based merge routine is invoked for embed_tokenslayer_range: [0, 1]
parameters:weight:0-sources:-model:cognitivecomputations/dolphin-2.2-70blayer_range: [1, 20]
-sources:-model:NousResearch/Nous-Hermes-2-Llama-2-70Blayer_range: [10, 30]
-sources:-model:cognitivecomputations/dolphin-2.2-70blayer_range: [20, 40]
-sources:-model:NousResearch/Nous-Hermes-2-Llama-2-70Blayer_range: [30, 50]
-sources:-model:cognitivecomputations/dolphin-2.2-70blayer_range: [40, 60]
-sources:-model:NousResearch/Nous-Hermes-2-Llama-2-70Blayer_range: [50, 70]
-sources:-model:cognitivecomputations/dolphin-2.2-70blayer_range: [60, 79]
-sources:# same as above, but for lm_head with the last layer-model:cognitivecomputations/dolphin-2.2-70blayer_range: [79, 80]
-model:NousResearch/Nous-Hermes-2-Llama-2-70Blayer_range: [79, 80]
parameters:weight:0dtype:float16tokenizer_source:model:cognitivecomputations/dolphin-2.2-70b# keep exact tokenizer used by dolphin - or you could use `union` if you add all of the input models to the first/last slice, but they would need to be non-zero weight or you'll get NaNs in your embeddings
Runs of dphn DolphinHermes-120b on huggingface.co
24
Total runs
-1
24-hour runs
0
3-day runs
0
7-day runs
7
30-day runs
More Information About DolphinHermes-120b huggingface.co Model
DolphinHermes-120b huggingface.co is an AI model on huggingface.co that provides DolphinHermes-120b's model effect (), which can be used instantly with this dphn DolphinHermes-120b model. huggingface.co supports a free trial of the DolphinHermes-120b model, and also provides paid use of the DolphinHermes-120b. Support call DolphinHermes-120b model through api, including Node.js, Python, http.
DolphinHermes-120b huggingface.co is an online trial and call api platform, which integrates DolphinHermes-120b's modeling effects, including api services, and provides a free online trial of DolphinHermes-120b, you can try DolphinHermes-120b online for free by clicking the link below.
dphn DolphinHermes-120b online free url in huggingface.co:
DolphinHermes-120b is an open source model from GitHub that offers a free installation service, and any user can find DolphinHermes-120b on GitHub to install. At the same time, huggingface.co provides the effect of DolphinHermes-120b install, users can directly use DolphinHermes-120b installed effect in huggingface.co for debugging and trial. It also supports api for free installation.