mergekit
mergekit copied to clipboard
[Bug?] `dict()` error when loading YAML config
I have this YAML config:
base_model: Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2
dtype: bfloat16
merge_method: breadcrumbs_ties
tokenizer:
source: union
parameters:
int8_mask: true
#normalize: true
random_seed: 145
models:
- model: arcee-ai/Llama-3.1-SuperNova-Lite
parameters:
weight:
- filter: lm_head
value: 0.0
- filter: self_attn.o_proj
value: 0.0
- filter: mlp.down_proj
value: 0.0
- value: 0.42
density: 0.9
gamma: 0.01
- model: VAGOsolutions/Llama-3.1-SauerkrautLM-8b-Instruct
parameters:
weight:
- filter: lm_head
value: 0.0
- filter: self_attn.o_proj
value: 0.0
- filter: mlp.down_proj
value: 0.0
- value: 0.33
density: 0.9
gamma: 0.01
- model: unsloth/Llama-3.1-Storm-8B
parameters:
weight:
- filter: lm_head
value: 0.0
- filter: self_attn.o_proj
value: 0.0
- filter: mlp.down_proj
value: 0.0
- value: 0.25
density: 0.9
gamma: 0.01
And am getting this error:
Traceback (most recent call last):
File "<frozen runpy>", line 198, in _run_module_as_main
File "<frozen runpy>", line 88, in _run_code
File "{ENV{\mergekit-yaml.exe\__main__.py", line 8, in <module>
File "{ENV}\click\core.py", line 1157, in __call__
return self.main(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "{ENV}\click\core.py", line 1078, in main
rv = self.invoke(ctx)
^^^^^^^^^^^^^^^^
File "{ENV}\click\core.py", line 1434, in invoke
return ctx.invoke(self.callback, **ctx.params)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "{ENV}\click\core.py", line 783, in invoke
return __callback(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "{ENV}\mergekit\options.py", line 82, in wrapper
f(*args, **kwargs)
File "{ENV}\mergekit\scripts\run_yaml.py", line 47, in main
run_merge(
File "{ENV}\mergekit\merge.py", line 96, in run_merge
for _task, value in exec.run(quiet=options.quiet):
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "{ENV}\mergekit\graph.py", line 197, in run
res = task.execute(**arguments)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "{ENV}\mergekit\tokenizer\embed.py", line 62, in execute
token_configs = dict(**self.tokens) or {}
^^^^^^^^^^^^^^^^^^^
TypeError: dict() argument after ** must be a mapping, not NoneType
What am I missing?