mradermacher

/

model_requests

like 115

This model unfortunately failed for with using the following error but worked for other quanters such as
bartowski: https://huggingface.co/bartowski/cerebras_GLM-4.5-Air-REAP-82B-A12B-GGUF - so probably fixable. Maybe @mradermacher knows more about how to fix this specific error. Maybe the model didn't download correctly?

GLM-4.5-Air-REAP-82B-A12B       INFO:hf-to-gguf:blk.46.nextn.eh_proj.weight,          torch.bfloat16 --> BF16, shape = {8192, 4096}
GLM-4.5-Air-REAP-82B-A12B       INFO:hf-to-gguf:blk.46.nextn.embed_tokens.weight,     torch.bfloat16 --> BF16, shape = {4096, 151552}
GLM-4.5-Air-REAP-82B-A12B       INFO:hf-to-gguf:blk.46.nextn.enorm.weight,            torch.bfloat16 --> F32, shape = {4096}
GLM-4.5-Air-REAP-82B-A12B       INFO:hf-to-gguf:blk.46.nextn.hnorm.weight,            torch.bfloat16 --> F32, shape = {4096}
GLM-4.5-Air-REAP-82B-A12B       INFO:hf-to-gguf:blk.46.attn_norm.weight,              torch.bfloat16 --> F32, shape = {4096}
GLM-4.5-Air-REAP-82B-A12B       Traceback (most recent call last):
GLM-4.5-Air-REAP-82B-A12B         File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 9554, in <module>
GLM-4.5-Air-REAP-82B-A12B           main()
GLM-4.5-Air-REAP-82B-A12B         File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 9548, in main
GLM-4.5-Air-REAP-82B-A12B           model_instance.write()
GLM-4.5-Air-REAP-82B-A12B         File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 442, in write
GLM-4.5-Air-REAP-82B-A12B           self.prepare_tensors()
GLM-4.5-Air-REAP-82B-A12B         File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 7341, in prepare_tensors
GLM-4.5-Air-REAP-82B-A12B           super().prepare_tensors()
GLM-4.5-Air-REAP-82B-A12B         File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 303, in prepare_tensors
GLM-4.5-Air-REAP-82B-A12B           for new_name, data_torch in (self.modify_tensors(data_torch, name, bid)):
GLM-4.5-Air-REAP-82B-A12B         File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 7320, in modify_tensors
GLM-4.5-Air-REAP-82B-A12B           datas.append(self._experts[bid][ename])
GLM-4.5-Air-REAP-82B-A12B       KeyError: 'model.layers.46.mlp.experts.8.down_proj.weight'
GLM-4.5-Air-REAP-82B-A12B       yes: standard output: Broken pipe

nope, no clue what that means. but i notice that some safetenbsors weer added later to that model, os maybe it failed because we tried the non-fixed versionj. i'll retry.

update: yeah, so, i guess that is a clue, let's hope it turns out to be correct :=)

yeah, seems to be working now