🌺 Ornith 1.0 Heretic 35B

#2719
by redaihf - opened

It's queued!

You can check for progress at http://hf.tst.eu/status.html or regularly check the model
summary page at https://hf.tst.eu/model#Ornith-1.0-35B-uncensored-heretic-GGUF for quants to appear.

Did something break here @RichardErkhov ?

Did something break here @RichardErkhov ?

Yes the converted GGUF unfortunately failed to run using llama.cpp:

Ornith-1.0-35B-uncensored-heretic       0.00.026.550 E ggml_cuda_init: failed to initialize CUDA: no CUDA-capable device is detected
Ornith-1.0-35B-uncensored-heretic       0.00.031.770 I llama_completion: llama backend init
Ornith-1.0-35B-uncensored-heretic       0.00.031.776 I llama_completion: load the model and apply lora adapter, if any
Ornith-1.0-35B-uncensored-heretic       0.00.118.974 W llama_model_loader: mmap is not supported for dry-run so it is now disabled
Ornith-1.0-35B-uncensored-heretic       0.00.451.050 E llama_model_load: error loading model: missing tensor 'blk.40.attn_norm.weight'
Ornith-1.0-35B-uncensored-heretic       0.00.451.078 E llama_model_load_from_file_impl: failed to load model
Ornith-1.0-35B-uncensored-heretic       0.00.451.085 E cmn  common_init_: failed to load model 'Ornith-1.0-35B-uncensored-heretic.gguf~'
Ornith-1.0-35B-uncensored-heretic       0.00.451.090 E llama_completion: error: unable to create context

llama_model_load: error loading model: missing tensor 'blk.40.attn_norm.weight'

Do you have any idea why @llmfan46 ?

llama_model_load: error loading model: missing tensor 'blk.40.attn_norm.weight'

Do you have any idea why @llmfan46 ?

No idea, you can see that I created GGUFs here:

https://huggingface.co/llmfan46/Ornith-1.0-35B-uncensored-heretic-GGUF

I am thinking it could either be a compatibility issue with different version of transformers (the safetensors where created with transformers 5.12.1) and/or a problem with the command line used, like maybe it needs --no-mtp at that end of the commend line since this model has no MTPs. I don't know, these are just guesses.

I will retry using latest llama.cpp and see if it still fails.

It’s now retrying using latest llama.cpp.

You can check for progress at http://hf.tst.eu/status.html or regularly check the model
summary page at https://hf.tst.eu/model#Ornith-1.0-35B-uncensored-heretic-GGUF for quants to appear.

Sign up or log in to comment