Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,68 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/_t60119d07h10m01,ep=2.0,mlr1.0e-03,b16,1,nb16,16,cL16,rR16,s1.0,inits43,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/_t60120d05h38m09,ep=2.0,mlr1.0e-03,b16,1,nb16,16,cL16,rR16,s1.0,inits44,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60111d08h06m53,ep=2.0,mlr2.0e-04,b8,nb4,4,cL4,rR4,s1.0,initdef,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60111d18h17m31,ep=2.0,mlr2.0e-04,b8,nb4,4,cL8,rR8,s2.0,initdef,dr0.0,size146627,2\*\*/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60116d00h39m41,ep=2.0,mlr4.0e-04,Lb16,1,nb32,32,cL32,rR32,s1.0,initlora32,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60117d14h18m29,ep=2.0,mlr4.0e-04,b16,1,nb32,32,cL32,rR32,s1.0,initLORA4,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60118d17h22m41,ep=2.0,mlr2.0e-04,b16,1,nb16,16,cL16,rR16,s1.0,initdef,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60119d00h34m38,ep=2.0,mlr2.0e-04,b16,1,nb4,4,cL4,rR4,s1.0,inits43,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60119d03h51m53,ep=2.0,mlr2.0e-04,b16,1,nb16,16,cL16,rR16,s1.0,inits43,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60119d10h52m09,ep=2.0,mlr1.0e-04,b16,1,nb32,32,cL32,rR32,s1.0,inits43,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60119d14h26m55,ep=2.0,mlr1.0e-04,b16,1,nb8,8,cL8,rR8,s1.0,inits43,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60119d19h46m19,ep=2.0,mlr1.0e-04,b16,1,nb4,4,cL4,rR4,s1.0,inits43,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60119d23h22m59,ep=2.0,mlr2.0e-04,b16,1,nb4,4,cL4,rR4,s1.0,inits44,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60120d02h28m13,ep=2.0,mlr2.0e-04,b16,1,nb16,16,cL16,rR16,s1.0,inits44,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60120d09h10m21,ep=2.0,mlr1.0e-04,b16,1,nb32,32,cL32,rR32,s1.0,inits44,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60120d12h41m43,ep=2.0,mlr1.0e-04,b16,1,nb8,8,cL8,rR8,s1.0,inits44,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60120d16h06m54,ep=2.0,mlr1.0e-04,b16,1,nb4,4,cL4,rR4,s1.0,inits44,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60120d23h33m42,ep=2.0,mlr2.0e-04,b16,1,nb4,4,cL4,rR4,s1.0,inits45,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60121d02h39m14,ep=2.0,mlr2.0e-04,b16,1,nb16,16,cL16,rR16,s1.0,inits45,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60121d05h44m56,ep=2.0,mlr1.0e-03,b16,1,nb16,16,cL16,rR16,s1.0,inits45,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60121d09h06m12,ep=2.0,mlr1.0e-04,b16,1,nb32,32,cL32,rR32,s1.0,inits45,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60121d12h32m06,ep=2.0,mlr1.0e-04,b16,1,nb8,8,cL8,rR8,s1.0,inits45,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60121d15h52m26,ep=2.0,mlr1.0e-04,b16,1,nb4,4,cL4,rR4,s1.0,inits45,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t60121d20h17m54,ep=2.0,mlr2.0e-04,b16,1,nb4,4,cL4,rR4,s1.0,initsd42,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t=31d04h17m34,ep=2.0,mlr1.0e-03,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t=31d11h50m33,ep=2.0,mlr1.0e-03,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t=60106d16h51m26,ep=2.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t=60106d21h14m30,ep=2.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size146627,6/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t=60107d01h46m41,ep=2.0,mlr2.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t=60107d06h16m57,ep=2.0,mlr1.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_B8/CMS/t=60107d10h45m00,ep=2.0,mlr2.0e-04,b8,nb16,16,cL16,rR16,s1,initdef,dr0.0,size146627,5/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_LoRA/CMS/t60115d00h54m28,ep=2.0,mlr1.0e-04,b16,1,nb32,32,cL32,rR32,s1.0,initlora32,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_LoRA/CMS/t60115d14h54m59,ep=2.0,mlr1.0e-04,b16,1,nb32,32,cL32,rR32,s1.0,initlora32,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_LoRA/CMS/t60116d00h05m29,ep=2.0,mlr1.0e-04,b16,1,nb32,32,cL32,rR32,s1.0,initlora32,dr0.0,size4000,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_LoRA/CMS/t60116d00h12m57,ep=2.0,mlr2.0e-04,b16,1,nb32,32,cL32,rR32,s1.0,initlora32,dr0.0,size4000,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_LoRA/CMS/t60116d00h20m28,ep=2.0,mlr5.0e-04,b16,1,nb32,32,cL32,rR32,s1.0,initlora32,dr0.0,size4000,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_LoRA/CMS/t60116d00h27m59,ep=2.0,mlr8.0e-04,b16,1,nb32,32,cL32,rR32,s1.0,initlora32,dr0.0,size4000,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama3_LoRA/CMS/t60116d03h20m53,ep=2.0,mlr7.0e-04,b16,1,nb32,32,cL32,rR32,s1.0,initlora32,dr0.0,size146627,2/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60112d11h56m00,mlr5.0e-04,b8,2,nb16,16,cL16,rR16,s1.0,initg9b,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60112d12h55m46,mlr5.0e-04,b8,2,nb4,4,cL4,rR4,s1.0,initg9b,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60112d14h14m09,mlr5.0e-04,b8,2,nb4,4,cL8,rR8,s2.0,initg9b,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60112d15h27m12,mlr5.0e-04,b8,2,nb4,4,cL16,rR16,s4.0,initg9b,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60112d17h16m48,mlr5.0e-04,b8,2,nb16,16,cL32,rR32,s2.0,initg9b,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60113d11h10m24,mlr5.0e-04,b8,2,nb2,2,cL2,rR2,s1.0,initg9b,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60113d11h52m59,mlr5.0e-04,b8,2,nb2,2,cL4,rR4,s2.0,initg9b,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60113d13h02m54,mlr5.0e-04,b8,2,nb2,2,cL4,rR4,s2.0,initg9b,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60123d00h51m06,mlr5.0e-04,b8,2,nb16,16,cL16,rR16,s1.0,initg9bs53,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60123d01h34m22,mlr5.0e-04,b8,2,nb2,2,cL2,rR2,s1.0,initg9bs53,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60123d02h16m48,mlr5.0e-04,b8,2,nb4,4,cL4,rR4,s1.0,initg9bs53,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60123d02h59m37,mlr5.0e-04,b8,2,nb2,2,cL4,rR4,s2.0,initg9bs53,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60123d03h42m41,mlr5.0e-04,b8,2,nb2,2,cL4,rR4,s1.4142,initg9bs53,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60123d04h25m42,mlr5.0e-04,b8,2,nb2,2,cL4,rR4,s2.8284,initg9bs53,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60123d05h08m42,mlr5.0e-04,b8,2,nb2,2,cL4,rR4,s4.0,initg9bs53,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60123d16h15m33,mlr5.0e-04,b8,2,nb2,2,cL4,rR4,s1.0,initg9bs53,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+MGemma9B/MATH/t60123d16h58m22,mlr5.0e-04,b8,2,nb2,2,cL2,rR2,s0.7071,initg9bs53,dr0.0,ep2.0/ft/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+dataset/hellaswag/test.json filter=lfs diff=lfs merge=lfs -text
+dataset/hellaswag/train.json filter=lfs diff=lfs merge=lfs -text
+dataset/social_i_qa/train.json filter=lfs diff=lfs merge=lfs -text
+dataset/winogrande/train.json filter=lfs diff=lfs merge=lfs -text
+ft_training_set/alpaca_data.json filter=lfs diff=lfs merge=lfs -text
+ft_training_set/alpaca_data_cleaned.json filter=lfs diff=lfs merge=lfs -text
+ft_training_set/commonsense_147k.json filter=lfs diff=lfs merge=lfs -text
+ft_training_set/commonsense_170k.json filter=lfs diff=lfs merge=lfs -text
+ft_training_set/math_14k.json filter=lfs diff=lfs merge=lfs -text
+ft_training_set/math_50k.json filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,237 @@
+wandb/*
+runs/*
+
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[codz]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#   Usually these files are written by a python script from a template
+#   before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py.cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+# Pipfile.lock
+
+# UV
+#   Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+# uv.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+# poetry.lock
+# poetry.toml
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#   pdm recommends including project-wide configuration in pdm.toml, but excluding .pdm-python.
+#   https://pdm-project.org/en/latest/usage/project/#working-with-version-control
+# pdm.lock
+# pdm.toml
+.pdm-python
+.pdm-build/
+
+# pixi
+#   Similar to Pipfile.lock, it is generally recommended to include pixi.lock in version control.
+# pixi.lock
+#   Pixi creates a virtual environment in the .pixi directory, just like venv module creates one
+#   in the .venv directory. It is recommended not to include this directory in version control.
+.pixi
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# Redis
+*.rdb
+*.aof
+*.pid
+
+# RabbitMQ
+mnesia/
+rabbitmq/
+rabbitmq-data/
+
+# ActiveMQ
+activemq-data/
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.envrc
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#   JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#   be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#   and can be added to the global gitignore or merged into this file.  For a more nuclear
+#   option (not recommended) you can uncomment the following to ignore the entire idea folder.
+# .idea/
+
+# Abstra
+#   Abstra is an AI-powered process automation framework.
+#   Ignore directories containing user credentials, local state, and settings.
+#   Learn more at https://abstra.io/docs
+.abstra/
+
+# Visual Studio Code
+#   Visual Studio Code specific template is maintained in a separate VisualStudioCode.gitignore 
+#   that can be found at https://github.com/github/gitignore/blob/main/Global/VisualStudioCode.gitignore
+#   and can be added to the global gitignore or merged into this file. However, if you prefer, 
+#   you could uncomment the following to ignore the entire vscode folder
+# .vscode/
+
+# Ruff stuff:
+.ruff_cache/
+
+# PyPI configuration file
+.pypirc
+
+# Marimo
+marimo/_static/
+marimo/_lsp/
+__marimo__/
+
+# Streamlit
+.streamlit/secrets.toml
+
+
+exps/
+wandb/
+# *.ipynb
+# scripts/all.sh
+# smpeft/sama/monarch_viz.png
+# Llama2_exps/
+# dataset/
+# ft_training_set/
+# Llama2_13B/
+# Llama13B/
+# Llama3_B00/
+# Llama3_B8/
+# Llama3_LoRA/
+# MGemma9B/
+# Mistral7B/

Llama13B/CMS/t60107d15h52m28,ep=2.0,mlr2.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "base_model_name_or_path": "dfurman/llama-13b",
+  "bias": "none",
+  "col_L": 32,
+  "drop_out": 0.0,
+  "inference_mode": false,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_unique_blocks_L": 32,
+  "num_unique_blocks_R": 32,
+  "peft_type": "SAMA",
+  "revision": null,
+  "row_R": 32,
+  "scaling": 1,
+  "share_factor_L": 1,
+  "share_factor_R": 1,
+  "target_modules": [
+    "k_proj",
+    "v_proj",
+    "q_proj",
+    "down_proj",
+    "up_proj"
+  ],
+  "target_modules_to_skip": null,
+  "task_type": "CAUSAL_LM"
+}

Llama13B/CMS/t60107d15h52m28,ep=2.0,mlr2.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

Llama13B/CMS/t60107d15h52m28,ep=2.0,mlr2.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

Llama13B/CMS/t60107d15h52m28,ep=2.0,mlr2.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "legacy": true,
+  "model_max_length": 512,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

Llama13B/CMS/t60107d15h52m28,ep=2.0,mlr2.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft2_/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "base_model_name_or_path": "dfurman/llama-13b",
+  "bias": "none",
+  "col_L": 32,
+  "drop_out": 0.0,
+  "inference_mode": true,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_unique_blocks_L": 32,
+  "num_unique_blocks_R": 32,
+  "peft_type": "SAMA",
+  "revision": null,
+  "row_R": 32,
+  "scaling": 1,
+  "share_factor_L": 1,
+  "share_factor_R": 1,
+  "target_modules": [
+    "k_proj",
+    "v_proj",
+    "q_proj",
+    "down_proj",
+    "up_proj"
+  ],
+  "target_modules_to_skip": null,
+  "task_type": "CAUSAL_LM"
+}

Llama13B/CMS/t60107d15h52m28,ep=2.0,mlr2.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft2_/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de64b12482a708fd656095a77f9647d214f1419ac5343723fd6acd16d2a8a391
+size 351427939

Llama13B/CMS/t60107d15h52m28,ep=2.0,mlr2.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/model_parameters_report.txt

@@ -0,0 +1,409 @@
+=== GLOBAL STATISTICS ===
+Total Parameters: 13,103,682,560
+Trainable Parameters: 87,818,240
+Trainable Percentage: 0.6702%
+
+================================================================================================================================
+=== DETAILED TRAINABLE MATRICES LIST ===
+Layer Name                                                                               | Shape                     | Count          
+---------------------------------------------------------------------------------------------------------------------------------------
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368

Llama13B/CMS/t60107d15h52m28,ep=2.0,mlr2.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/trainer_state.json

@@ -0,0 +1,110 @@
+{
+  "best_global_step": 1500,
+  "best_metric": 0.05616725981235504,
+  "best_model_checkpoint": "./Llama13B/CMS/t60107d15h52m28,ep=2.0,mlr2.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/checkpoint-1500",
+  "epoch": 2.0,
+  "eval_steps": 300,
+  "global_step": 1766,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.33994334277620397,
+      "grad_norm": 0.07389352470636368,
+      "learning_rate": 0.0001930248326634556,
+      "loss": 0.5654,
+      "step": 300
+    },
+    {
+      "epoch": 0.6798866855524079,
+      "grad_norm": 0.07417800277471542,
+      "learning_rate": 0.0001588090691373661,
+      "loss": 0.0754,
+      "step": 600
+    },
+    {
+      "epoch": 0.6798866855524079,
+      "eval_loss": 0.06705917418003082,
+      "eval_runtime": 43.1442,
+      "eval_samples_per_second": 23.178,
+      "eval_steps_per_second": 0.742,
+      "step": 600
+    },
+    {
+      "epoch": 1.0192634560906515,
+      "grad_norm": 0.0427674800157547,
+      "learning_rate": 0.00010622629205383885,
+      "loss": 0.062,
+      "step": 900
+    },
+    {
+      "epoch": 1.0192634560906515,
+      "eval_loss": 0.060582421720027924,
+      "eval_runtime": 42.6357,
+      "eval_samples_per_second": 23.455,
+      "eval_steps_per_second": 0.751,
+      "step": 900
+    },
+    {
+      "epoch": 1.3592067988668555,
+      "grad_norm": 0.05241871252655983,
+      "learning_rate": 5.1698944307268006e-05,
+      "loss": 0.0483,
+      "step": 1200
+    },
+    {
+      "epoch": 1.3592067988668555,
+      "eval_loss": 0.05789685994386673,
+      "eval_runtime": 42.8479,
+      "eval_samples_per_second": 23.338,
+      "eval_steps_per_second": 0.747,
+      "step": 1200
+    },
+    {
+      "epoch": 1.6991501416430594,
+      "grad_norm": 0.07206667214632034,
+      "learning_rate": 1.225678933385207e-05,
+      "loss": 0.046,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6991501416430594,
+      "eval_loss": 0.05616725981235504,
+      "eval_runtime": 42.8072,
+      "eval_samples_per_second": 23.361,
+      "eval_steps_per_second": 0.748,
+      "step": 1500
+    },
+    {
+      "epoch": 2.0,
+      "step": 1766,
+      "total_flos": 1.1224942823709082e+18,
+      "train_loss": 0.14243490050630148,
+      "train_runtime": 3020.9682,
+      "train_samples_per_second": 9.347,
+      "train_steps_per_second": 0.585
+    }
+  ],
+  "logging_steps": 300,
+  "max_steps": 1766,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 300,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.1224942823709082e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

Llama13B/CMS/t60107d17h56m56,ep=2.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "base_model_name_or_path": "dfurman/llama-13b",
+  "bias": "none",
+  "col_L": 32,
+  "drop_out": 0.0,
+  "inference_mode": false,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_unique_blocks_L": 32,
+  "num_unique_blocks_R": 32,
+  "peft_type": "SAMA",
+  "revision": null,
+  "row_R": 32,
+  "scaling": 1,
+  "share_factor_L": 1,
+  "share_factor_R": 1,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "down_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "target_modules_to_skip": null,
+  "task_type": "CAUSAL_LM"
+}

Llama13B/CMS/t60107d17h56m56,ep=2.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

Llama13B/CMS/t60107d17h56m56,ep=2.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

Llama13B/CMS/t60107d17h56m56,ep=2.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "legacy": true,
+  "model_max_length": 512,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

Llama13B/CMS/t60107d17h56m56,ep=2.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft2_/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "base_model_name_or_path": "dfurman/llama-13b",
+  "bias": "none",
+  "col_L": 32,
+  "drop_out": 0.0,
+  "inference_mode": true,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_unique_blocks_L": 32,
+  "num_unique_blocks_R": 32,
+  "peft_type": "SAMA",
+  "revision": null,
+  "row_R": 32,
+  "scaling": 1,
+  "share_factor_L": 1,
+  "share_factor_R": 1,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "down_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "target_modules_to_skip": null,
+  "task_type": "CAUSAL_LM"
+}

Llama13B/CMS/t60107d17h56m56,ep=2.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft2_/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab7a6902cdb949231f188dd007f69b68b4bdcab0612a9c5025026c153f0b0cfb
+size 351427939

Llama13B/CMS/t60107d17h56m56,ep=2.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/model_parameters_report.txt

@@ -0,0 +1,409 @@
+=== GLOBAL STATISTICS ===
+Total Parameters: 13,103,682,560
+Trainable Parameters: 87,818,240
+Trainable Percentage: 0.6702%
+
+================================================================================================================================
+=== DETAILED TRAINABLE MATRICES LIST ===
+Layer Name                                                                               | Shape                     | Count          
+---------------------------------------------------------------------------------------------------------------------------------------
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368

Llama13B/CMS/t60107d17h56m56,ep=2.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/trainer_state.json

@@ -0,0 +1,528 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 50,
+  "global_step": 1766,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.056657223796033995,
+      "grad_norm": 0.03540075570344925,
+      "learning_rate": 0.000245,
+      "loss": 2.0119,
+      "step": 50
+    },
+    {
+      "epoch": 0.11331444759206799,
+      "grad_norm": 0.08595774322748184,
+      "learning_rate": 0.000495,
+      "loss": 0.1096,
+      "step": 100
+    },
+    {
+      "epoch": 0.16997167138810199,
+      "grad_norm": 0.0386439748108387,
+      "learning_rate": 0.0004989309807668432,
+      "loss": 0.0948,
+      "step": 150
+    },
+    {
+      "epoch": 0.22662889518413598,
+      "grad_norm": 0.08770483732223511,
+      "learning_rate": 0.000495645795847958,
+      "loss": 0.0867,
+      "step": 200
+    },
+    {
+      "epoch": 0.28328611898017,
+      "grad_norm": 0.08420336246490479,
+      "learning_rate": 0.0004901732526860121,
+      "loss": 0.075,
+      "step": 250
+    },
+    {
+      "epoch": 0.33994334277620397,
+      "grad_norm": 0.1704639345407486,
+      "learning_rate": 0.000482562081658639,
+      "loss": 0.0667,
+      "step": 300
+    },
+    {
+      "epoch": 0.33994334277620397,
+      "eval_loss": 0.07056522369384766,
+      "eval_runtime": 43.0303,
+      "eval_samples_per_second": 23.239,
+      "eval_steps_per_second": 0.744,
+      "step": 300
+    },
+    {
+      "epoch": 0.39660056657223797,
+      "grad_norm": 0.09642501175403595,
+      "learning_rate": 0.00047288005660104163,
+      "loss": 0.0671,
+      "step": 350
+    },
+    {
+      "epoch": 0.39660056657223797,
+      "eval_loss": 0.0681549534201622,
+      "eval_runtime": 42.7781,
+      "eval_samples_per_second": 23.376,
+      "eval_steps_per_second": 0.748,
+      "step": 350
+    },
+    {
+      "epoch": 0.45325779036827196,
+      "grad_norm": 0.05195102468132973,
+      "learning_rate": 0.0004612133913124268,
+      "loss": 0.0604,
+      "step": 400
+    },
+    {
+      "epoch": 0.45325779036827196,
+      "eval_loss": 0.06310559809207916,
+      "eval_runtime": 42.7964,
+      "eval_samples_per_second": 23.366,
+      "eval_steps_per_second": 0.748,
+      "step": 400
+    },
+    {
+      "epoch": 0.509915014164306,
+      "grad_norm": 0.0997244343161583,
+      "learning_rate": 0.0004476659718633865,
+      "loss": 0.0596,
+      "step": 450
+    },
+    {
+      "epoch": 0.509915014164306,
+      "eval_loss": 0.06353795528411865,
+      "eval_runtime": 42.8058,
+      "eval_samples_per_second": 23.361,
+      "eval_steps_per_second": 0.748,
+      "step": 450
+    },
+    {
+      "epoch": 0.56657223796034,
+      "grad_norm": 0.03704178333282471,
+      "learning_rate": 0.0004323584315401604,
+      "loss": 0.059,
+      "step": 500
+    },
+    {
+      "epoch": 0.56657223796034,
+      "eval_loss": 0.06257344037294388,
+      "eval_runtime": 42.7483,
+      "eval_samples_per_second": 23.393,
+      "eval_steps_per_second": 0.749,
+      "step": 500
+    },
+    {
+      "epoch": 0.623229461756374,
+      "grad_norm": 0.10769298672676086,
+      "learning_rate": 0.000415427076662958,
+      "loss": 0.0575,
+      "step": 550
+    },
+    {
+      "epoch": 0.623229461756374,
+      "eval_loss": 0.05993322283029556,
+      "eval_runtime": 42.7557,
+      "eval_samples_per_second": 23.389,
+      "eval_steps_per_second": 0.748,
+      "step": 550
+    },
+    {
+      "epoch": 0.6798866855524079,
+      "grad_norm": 0.08261393010616302,
+      "learning_rate": 0.0003970226728434152,
+      "loss": 0.0611,
+      "step": 600
+    },
+    {
+      "epoch": 0.6798866855524079,
+      "eval_loss": 0.05639559403061867,
+      "eval_runtime": 42.7842,
+      "eval_samples_per_second": 23.373,
+      "eval_steps_per_second": 0.748,
+      "step": 600
+    },
+    {
+      "epoch": 0.7365439093484419,
+      "grad_norm": 0.03413009271025658,
+      "learning_rate": 0.00037730910248898127,
+      "loss": 0.0566,
+      "step": 650
+    },
+    {
+      "epoch": 0.7365439093484419,
+      "eval_loss": 0.055478040128946304,
+      "eval_runtime": 42.7805,
+      "eval_samples_per_second": 23.375,
+      "eval_steps_per_second": 0.748,
+      "step": 650
+    },
+    {
+      "epoch": 0.7932011331444759,
+      "grad_norm": 0.14471058547496796,
+      "learning_rate": 0.00035646190550851575,
+      "loss": 0.0549,
+      "step": 700
+    },
+    {
+      "epoch": 0.7932011331444759,
+      "eval_loss": 0.05561484023928642,
+      "eval_runtime": 42.7721,
+      "eval_samples_per_second": 23.38,
+      "eval_steps_per_second": 0.748,
+      "step": 700
+    },
+    {
+      "epoch": 0.8498583569405099,
+      "grad_norm": 0.08413548767566681,
+      "learning_rate": 0.0003346667162134164,
+      "loss": 0.0555,
+      "step": 750
+    },
+    {
+      "epoch": 0.8498583569405099,
+      "eval_loss": 0.053136855363845825,
+      "eval_runtime": 42.7831,
+      "eval_samples_per_second": 23.374,
+      "eval_steps_per_second": 0.748,
+      "step": 750
+    },
+    {
+      "epoch": 0.9065155807365439,
+      "grad_norm": 0.06495863944292068,
+      "learning_rate": 0.00031211761033292194,
+      "loss": 0.0516,
+      "step": 800
+    },
+    {
+      "epoch": 0.9065155807365439,
+      "eval_loss": 0.051497943699359894,
+      "eval_runtime": 42.8012,
+      "eval_samples_per_second": 23.364,
+      "eval_steps_per_second": 0.748,
+      "step": 800
+    },
+    {
+      "epoch": 0.9631728045325779,
+      "grad_norm": 0.06451834738254547,
+      "learning_rate": 0.0002890153768626315,
+      "loss": 0.0533,
+      "step": 850
+    },
+    {
+      "epoch": 0.9631728045325779,
+      "eval_loss": 0.0497167594730854,
+      "eval_runtime": 42.7769,
+      "eval_samples_per_second": 23.377,
+      "eval_steps_per_second": 0.748,
+      "step": 850
+    },
+    {
+      "epoch": 1.0192634560906515,
+      "grad_norm": 0.023159746080636978,
+      "learning_rate": 0.0002655657301345971,
+      "loss": 0.0446,
+      "step": 900
+    },
+    {
+      "epoch": 1.0192634560906515,
+      "eval_loss": 0.05106104165315628,
+      "eval_runtime": 42.7088,
+      "eval_samples_per_second": 23.414,
+      "eval_steps_per_second": 0.749,
+      "step": 900
+    },
+    {
+      "epoch": 1.0759206798866856,
+      "grad_norm": 0.0639742985367775,
+      "learning_rate": 0.00024197747802965556,
+      "loss": 0.0309,
+      "step": 950
+    },
+    {
+      "epoch": 1.0759206798866856,
+      "eval_loss": 0.04969682916998863,
+      "eval_runtime": 42.781,
+      "eval_samples_per_second": 23.375,
+      "eval_steps_per_second": 0.748,
+      "step": 950
+    },
+    {
+      "epoch": 1.1325779036827195,
+      "grad_norm": 0.048771053552627563,
+      "learning_rate": 0.00021846066264318665,
+      "loss": 0.0318,
+      "step": 1000
+    },
+    {
+      "epoch": 1.1325779036827195,
+      "eval_loss": 0.05130666494369507,
+      "eval_runtime": 42.7991,
+      "eval_samples_per_second": 23.365,
+      "eval_steps_per_second": 0.748,
+      "step": 1000
+    },
+    {
+      "epoch": 1.1892351274787536,
+      "grad_norm": 0.07568787038326263,
+      "learning_rate": 0.00019522468996078257,
+      "loss": 0.0333,
+      "step": 1050
+    },
+    {
+      "epoch": 1.1892351274787536,
+      "eval_loss": 0.05067039653658867,
+      "eval_runtime": 42.8009,
+      "eval_samples_per_second": 23.364,
+      "eval_steps_per_second": 0.748,
+      "step": 1050
+    },
+    {
+      "epoch": 1.2458923512747875,
+      "grad_norm": 0.08877400308847427,
+      "learning_rate": 0.00017247646519817056,
+      "loss": 0.0333,
+      "step": 1100
+    },
+    {
+      "epoch": 1.2458923512747875,
+      "eval_loss": 0.05075615644454956,
+      "eval_runtime": 42.7716,
+      "eval_samples_per_second": 23.38,
+      "eval_steps_per_second": 0.748,
+      "step": 1100
+    },
+    {
+      "epoch": 1.3025495750708216,
+      "grad_norm": 0.04937183856964111,
+      "learning_rate": 0.00015041855040929373,
+      "loss": 0.0326,
+      "step": 1150
+    },
+    {
+      "epoch": 1.3025495750708216,
+      "eval_loss": 0.05133920907974243,
+      "eval_runtime": 42.7949,
+      "eval_samples_per_second": 23.367,
+      "eval_steps_per_second": 0.748,
+      "step": 1150
+    },
+    {
+      "epoch": 1.3592067988668555,
+      "grad_norm": 0.09826551377773285,
+      "learning_rate": 0.00012924736076817,
+      "loss": 0.0283,
+      "step": 1200
+    },
+    {
+      "epoch": 1.3592067988668555,
+      "eval_loss": 0.05426332727074623,
+      "eval_runtime": 42.7975,
+      "eval_samples_per_second": 23.366,
+      "eval_steps_per_second": 0.748,
+      "step": 1200
+    },
+    {
+      "epoch": 1.4158640226628896,
+      "grad_norm": 0.08305229991674423,
+      "learning_rate": 0.00010915141558577307,
+      "loss": 0.0342,
+      "step": 1250
+    },
+    {
+      "epoch": 1.4158640226628896,
+      "eval_loss": 0.050987694412469864,
+      "eval_runtime": 42.8194,
+      "eval_samples_per_second": 23.354,
+      "eval_steps_per_second": 0.747,
+      "step": 1250
+    },
+    {
+      "epoch": 1.4725212464589235,
+      "grad_norm": 0.04723089560866356,
+      "learning_rate": 9.030965963579688e-05,
+      "loss": 0.0293,
+      "step": 1300
+    },
+    {
+      "epoch": 1.4725212464589235,
+      "eval_loss": 0.04957663640379906,
+      "eval_runtime": 42.8181,
+      "eval_samples_per_second": 23.355,
+      "eval_steps_per_second": 0.747,
+      "step": 1300
+    },
+    {
+      "epoch": 1.5291784702549576,
+      "grad_norm": 0.03975960984826088,
+      "learning_rate": 7.288986973708988e-05,
+      "loss": 0.0317,
+      "step": 1350
+    },
+    {
+      "epoch": 1.5291784702549576,
+      "eval_loss": 0.0498320609331131,
+      "eval_runtime": 42.7964,
+      "eval_samples_per_second": 23.366,
+      "eval_steps_per_second": 0.748,
+      "step": 1350
+    },
+    {
+      "epoch": 1.5858356940509915,
+      "grad_norm": 0.10585972666740417,
+      "learning_rate": 5.704716078138403e-05,
+      "loss": 0.0329,
+      "step": 1400
+    },
+    {
+      "epoch": 1.5858356940509915,
+      "eval_loss": 0.0496598519384861,
+      "eval_runtime": 42.8107,
+      "eval_samples_per_second": 23.359,
+      "eval_steps_per_second": 0.747,
+      "step": 1400
+    },
+    {
+      "epoch": 1.6424929178470253,
+      "grad_norm": 0.09209132194519043,
+      "learning_rate": 4.2922604509423035e-05,
+      "loss": 0.0245,
+      "step": 1450
+    },
+    {
+      "epoch": 1.6424929178470253,
+      "eval_loss": 0.050136879086494446,
+      "eval_runtime": 42.7941,
+      "eval_samples_per_second": 23.368,
+      "eval_steps_per_second": 0.748,
+      "step": 1450
+    },
+    {
+      "epoch": 1.6991501416430594,
+      "grad_norm": 0.05146324634552002,
+      "learning_rate": 3.064197333463017e-05,
+      "loss": 0.0299,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6991501416430594,
+      "eval_loss": 0.05017969012260437,
+      "eval_runtime": 42.8143,
+      "eval_samples_per_second": 23.357,
+      "eval_steps_per_second": 0.747,
+      "step": 1500
+    },
+    {
+      "epoch": 1.7558073654390935,
+      "grad_norm": 0.06396601349115372,
+      "learning_rate": 2.0314620399962458e-05,
+      "loss": 0.0305,
+      "step": 1550
+    },
+    {
+      "epoch": 1.7558073654390935,
+      "eval_loss": 0.049846261739730835,
+      "eval_runtime": 42.801,
+      "eval_samples_per_second": 23.364,
+      "eval_steps_per_second": 0.748,
+      "step": 1550
+    },
+    {
+      "epoch": 1.8124645892351274,
+      "grad_norm": 0.07966850697994232,
+      "learning_rate": 1.2032505840510632e-05,
+      "loss": 0.0306,
+      "step": 1600
+    },
+    {
+      "epoch": 1.8124645892351274,
+      "eval_loss": 0.04927060008049011,
+      "eval_runtime": 42.7981,
+      "eval_samples_per_second": 23.366,
+      "eval_steps_per_second": 0.748,
+      "step": 1600
+    },
+    {
+      "epoch": 1.8691218130311613,
+      "grad_norm": 0.03877720236778259,
+      "learning_rate": 5.869377922509161e-06,
+      "loss": 0.0299,
+      "step": 1650
+    },
+    {
+      "epoch": 1.8691218130311613,
+      "eval_loss": 0.04900398850440979,
+      "eval_runtime": 42.7902,
+      "eval_samples_per_second": 23.37,
+      "eval_steps_per_second": 0.748,
+      "step": 1650
+    },
+    {
+      "epoch": 1.9257790368271954,
+      "grad_norm": 0.016926735639572144,
+      "learning_rate": 1.8801163503225082e-06,
+      "loss": 0.0321,
+      "step": 1700
+    },
+    {
+      "epoch": 1.9257790368271954,
+      "eval_loss": 0.04912006855010986,
+      "eval_runtime": 42.8004,
+      "eval_samples_per_second": 23.364,
+      "eval_steps_per_second": 0.748,
+      "step": 1700
+    },
+    {
+      "epoch": 1.9824362606232295,
+      "grad_norm": 0.09555891156196594,
+      "learning_rate": 1.0024358894342167e-07,
+      "loss": 0.0307,
+      "step": 1750
+    },
+    {
+      "epoch": 1.9824362606232295,
+      "eval_loss": 0.04891112446784973,
+      "eval_runtime": 42.7732,
+      "eval_samples_per_second": 23.379,
+      "eval_steps_per_second": 0.748,
+      "step": 1750
+    },
+    {
+      "epoch": 2.0,
+      "step": 1766,
+      "total_flos": 1.1224942823709082e+18,
+      "train_loss": 0.10363658229870003,
+      "train_runtime": 2902.2639,
+      "train_samples_per_second": 9.73,
+      "train_steps_per_second": 0.608
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 1766,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 50.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.1224942823709082e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

Llama13B/CMS/t60107d20h20m59,ep=1.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "base_model_name_or_path": "dfurman/llama-13b",
+  "bias": "none",
+  "col_L": 32,
+  "drop_out": 0.0,
+  "inference_mode": false,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_unique_blocks_L": 32,
+  "num_unique_blocks_R": 32,
+  "peft_type": "SAMA",
+  "revision": null,
+  "row_R": 32,
+  "scaling": 1,
+  "share_factor_L": 1,
+  "share_factor_R": 1,
+  "target_modules": [
+    "down_proj",
+    "up_proj",
+    "q_proj",
+    "v_proj",
+    "k_proj"
+  ],
+  "target_modules_to_skip": null,
+  "task_type": "CAUSAL_LM"
+}

Llama13B/CMS/t60107d20h20m59,ep=1.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

Llama13B/CMS/t60107d20h20m59,ep=1.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

Llama13B/CMS/t60107d20h20m59,ep=1.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "legacy": true,
+  "model_max_length": 512,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

Llama13B/CMS/t60107d20h20m59,ep=1.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft2_/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "base_model_name_or_path": "dfurman/llama-13b",
+  "bias": "none",
+  "col_L": 32,
+  "drop_out": 0.0,
+  "inference_mode": true,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_unique_blocks_L": 32,
+  "num_unique_blocks_R": 32,
+  "peft_type": "SAMA",
+  "revision": null,
+  "row_R": 32,
+  "scaling": 1,
+  "share_factor_L": 1,
+  "share_factor_R": 1,
+  "target_modules": [
+    "down_proj",
+    "up_proj",
+    "q_proj",
+    "v_proj",
+    "k_proj"
+  ],
+  "target_modules_to_skip": null,
+  "task_type": "CAUSAL_LM"
+}

Llama13B/CMS/t60107d20h20m59,ep=1.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/ft2_/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63407e24c9c8aa2349eff68dd62b89733eea72c3b8e4de0a7bf654970c28ba41
+size 351427939

Llama13B/CMS/t60107d20h20m59,ep=1.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/model_parameters_report.txt

@@ -0,0 +1,409 @@
+=== GLOBAL STATISTICS ===
+Total Parameters: 13,103,682,560
+Trainable Parameters: 87,818,240
+Trainable Percentage: 0.6702%
+
+================================================================================================================================
+=== DETAILED TRAINABLE MATRICES LIST ===
+Layer Name                                                                               | Shape                     | Count          
+---------------------------------------------------------------------------------------------------------------------------------------
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_R.weights      | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 32]             | 442,368
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_R.weights           | [32, 32, 160]             | 163,840
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 32]             | 163,840
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_R.weights         | [32, 32, 432]             | 442,368
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_R.weights     | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 32]             | 442,368
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_R.weights          | [32, 32, 160]             | 163,840
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 32]             | 163,840
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_R.weights        | [32, 32, 432]             | 442,368

Llama13B/CMS/t60107d20h20m59,ep=1.0,mlr5.0e-04,b8,nb32,32,cL32,rR32,s1,initdef,dr0.0,size14119,5/trainer_state.json

@@ -0,0 +1,528 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 50,
+  "global_step": 1765,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.028328611898016998,
+      "grad_norm": 0.05120686814188957,
+      "learning_rate": 0.000245,
+      "loss": 1.904,
+      "step": 50
+    },
+    {
+      "epoch": 0.056657223796033995,
+      "grad_norm": 0.0301814004778862,
+      "learning_rate": 0.000495,
+      "loss": 0.1162,
+      "step": 100
+    },
+    {
+      "epoch": 0.08498583569405099,
+      "grad_norm": 0.05142534524202347,
+      "learning_rate": 0.0004989309807668432,
+      "loss": 0.0918,
+      "step": 150
+    },
+    {
+      "epoch": 0.11331444759206799,
+      "grad_norm": 0.04382554814219475,
+      "learning_rate": 0.000495645795847958,
+      "loss": 0.0934,
+      "step": 200
+    },
+    {
+      "epoch": 0.141643059490085,
+      "grad_norm": 0.07652860879898071,
+      "learning_rate": 0.0004901732526860121,
+      "loss": 0.0842,
+      "step": 250
+    },
+    {
+      "epoch": 0.16997167138810199,
+      "grad_norm": 0.08357395231723785,
+      "learning_rate": 0.000482562081658639,
+      "loss": 0.0769,
+      "step": 300
+    },
+    {
+      "epoch": 0.16997167138810199,
+      "eval_loss": 0.08110018074512482,
+      "eval_runtime": 40.2393,
+      "eval_samples_per_second": 24.851,
+      "eval_steps_per_second": 0.795,
+      "step": 300
+    },
+    {
+      "epoch": 0.19830028328611898,
+      "grad_norm": 0.09154224395751953,
+      "learning_rate": 0.00047288005660104163,
+      "loss": 0.08,
+      "step": 350
+    },
+    {
+      "epoch": 0.19830028328611898,
+      "eval_loss": 0.07602168619632721,
+      "eval_runtime": 40.0059,
+      "eval_samples_per_second": 24.996,
+      "eval_steps_per_second": 0.8,
+      "step": 350
+    },
+    {
+      "epoch": 0.22662889518413598,
+      "grad_norm": 0.06302578002214432,
+      "learning_rate": 0.0004612133913124268,
+      "loss": 0.0739,
+      "step": 400
+    },
+    {
+      "epoch": 0.22662889518413598,
+      "eval_loss": 0.06869283318519592,
+      "eval_runtime": 40.0279,
+      "eval_samples_per_second": 24.983,
+      "eval_steps_per_second": 0.799,
+      "step": 400
+    },
+    {
+      "epoch": 0.254957507082153,
+      "grad_norm": 0.21662577986717224,
+      "learning_rate": 0.0004476659718633865,
+      "loss": 0.0707,
+      "step": 450
+    },
+    {
+      "epoch": 0.254957507082153,
+      "eval_loss": 0.06703159958124161,
+      "eval_runtime": 40.0265,
+      "eval_samples_per_second": 24.983,
+      "eval_steps_per_second": 0.799,
+      "step": 450
+    },
+    {
+      "epoch": 0.28328611898017,
+      "grad_norm": 0.0684773176908493,
+      "learning_rate": 0.0004323584315401604,
+      "loss": 0.0666,
+      "step": 500
+    },
+    {
+      "epoch": 0.28328611898017,
+      "eval_loss": 0.06985477358102798,
+      "eval_runtime": 39.9964,
+      "eval_samples_per_second": 25.002,
+      "eval_steps_per_second": 0.8,
+      "step": 500
+    },
+    {
+      "epoch": 0.311614730878187,
+      "grad_norm": 0.023050207644701004,
+      "learning_rate": 0.000415427076662958,
+      "loss": 0.065,
+      "step": 550
+    },
+    {
+      "epoch": 0.311614730878187,
+      "eval_loss": 0.06582478433847427,
+      "eval_runtime": 39.9999,
+      "eval_samples_per_second": 25.0,
+      "eval_steps_per_second": 0.8,
+      "step": 550
+    },
+    {
+      "epoch": 0.33994334277620397,
+      "grad_norm": 0.06662499159574509,
+      "learning_rate": 0.0003970226728434152,
+      "loss": 0.0627,
+      "step": 600
+    },
+    {
+      "epoch": 0.33994334277620397,
+      "eval_loss": 0.06325174123048782,
+      "eval_runtime": 39.9867,
+      "eval_samples_per_second": 25.008,
+      "eval_steps_per_second": 0.8,
+      "step": 600
+    },
+    {
+      "epoch": 0.36827195467422097,
+      "grad_norm": 0.06945322453975677,
+      "learning_rate": 0.00037730910248898127,
+      "loss": 0.0661,
+      "step": 650
+    },
+    {
+      "epoch": 0.36827195467422097,
+      "eval_loss": 0.060704972594976425,
+      "eval_runtime": 39.9924,
+      "eval_samples_per_second": 25.005,
+      "eval_steps_per_second": 0.8,
+      "step": 650
+    },
+    {
+      "epoch": 0.39660056657223797,
+      "grad_norm": 0.21134211122989655,
+      "learning_rate": 0.00035646190550851575,
+      "loss": 0.061,
+      "step": 700
+    },
+    {
+      "epoch": 0.39660056657223797,
+      "eval_loss": 0.05985141918063164,
+      "eval_runtime": 40.0163,
+      "eval_samples_per_second": 24.99,
+      "eval_steps_per_second": 0.8,
+      "step": 700
+    },
+    {
+      "epoch": 0.42492917847025496,
+      "grad_norm": 0.059777744114398956,
+      "learning_rate": 0.0003346667162134164,
+      "loss": 0.0556,
+      "step": 750
+    },
+    {
+      "epoch": 0.42492917847025496,
+      "eval_loss": 0.06151320040225983,
+      "eval_runtime": 40.0008,
+      "eval_samples_per_second": 25.0,
+      "eval_steps_per_second": 0.8,
+      "step": 750
+    },
+    {
+      "epoch": 0.45325779036827196,
+      "grad_norm": 0.0928986445069313,
+      "learning_rate": 0.00031211761033292194,
+      "loss": 0.0566,
+      "step": 800
+    },
+    {
+      "epoch": 0.45325779036827196,
+      "eval_loss": 0.05747922882437706,
+      "eval_runtime": 40.0033,
+      "eval_samples_per_second": 24.998,
+      "eval_steps_per_second": 0.8,
+      "step": 800
+    },
+    {
+      "epoch": 0.48158640226628896,
+      "grad_norm": 0.12957215309143066,
+      "learning_rate": 0.0002890153768626315,
+      "loss": 0.059,
+      "step": 850
+    },
+    {
+      "epoch": 0.48158640226628896,
+      "eval_loss": 0.057896047830581665,
+      "eval_runtime": 40.0121,
+      "eval_samples_per_second": 24.992,
+      "eval_steps_per_second": 0.8,
+      "step": 850
+    },
+    {
+      "epoch": 0.509915014164306,
+      "grad_norm": 0.1764892339706421,
+      "learning_rate": 0.0002655657301345971,
+      "loss": 0.0539,
+      "step": 900
+    },
+    {
+      "epoch": 0.509915014164306,
+      "eval_loss": 0.05976394563913345,
+      "eval_runtime": 40.0113,
+      "eval_samples_per_second": 24.993,
+      "eval_steps_per_second": 0.8,
+      "step": 900
+    },
+    {
+      "epoch": 0.5382436260623229,
+      "grad_norm": 0.05985495075583458,
+      "learning_rate": 0.00024197747802965556,
+      "loss": 0.0558,
+      "step": 950
+    },
+    {
+      "epoch": 0.5382436260623229,
+      "eval_loss": 0.053400635719299316,
+      "eval_runtime": 39.9933,
+      "eval_samples_per_second": 25.004,
+      "eval_steps_per_second": 0.8,
+      "step": 950
+    },
+    {
+      "epoch": 0.56657223796034,
+      "grad_norm": 0.06537602096796036,
+      "learning_rate": 0.00021846066264318665,
+      "loss": 0.0509,
+      "step": 1000
+    },
+    {
+      "epoch": 0.56657223796034,
+      "eval_loss": 0.05671335384249687,
+      "eval_runtime": 40.0061,
+      "eval_samples_per_second": 24.996,
+      "eval_steps_per_second": 0.8,
+      "step": 1000
+    },
+    {
+      "epoch": 0.5949008498583569,
+      "grad_norm": 0.06947670876979828,
+      "learning_rate": 0.00019522468996078257,
+      "loss": 0.053,
+      "step": 1050
+    },
+    {
+      "epoch": 0.5949008498583569,
+      "eval_loss": 0.05418495461344719,
+      "eval_runtime": 39.9757,
+      "eval_samples_per_second": 25.015,
+      "eval_steps_per_second": 0.8,
+      "step": 1050
+    },
+    {
+      "epoch": 0.623229461756374,
+      "grad_norm": 0.2086954563856125,
+      "learning_rate": 0.00017247646519817056,
+      "loss": 0.0502,
+      "step": 1100
+    },
+    {
+      "epoch": 0.623229461756374,
+      "eval_loss": 0.05340389162302017,
+      "eval_runtime": 39.9938,
+      "eval_samples_per_second": 25.004,
+      "eval_steps_per_second": 0.8,
+      "step": 1100
+    },
+    {
+      "epoch": 0.6515580736543909,
+      "grad_norm": 0.11571495234966278,
+      "learning_rate": 0.00015041855040929373,
+      "loss": 0.0523,
+      "step": 1150
+    },
+    {
+      "epoch": 0.6515580736543909,
+      "eval_loss": 0.05297320336103439,
+      "eval_runtime": 39.9981,
+      "eval_samples_per_second": 25.001,
+      "eval_steps_per_second": 0.8,
+      "step": 1150
+    },
+    {
+      "epoch": 0.6798866855524079,
+      "grad_norm": 0.07803928852081299,
+      "learning_rate": 0.00012924736076817,
+      "loss": 0.0555,
+      "step": 1200
+    },
+    {
+      "epoch": 0.6798866855524079,
+      "eval_loss": 0.0508769154548645,
+      "eval_runtime": 39.9981,
+      "eval_samples_per_second": 25.001,
+      "eval_steps_per_second": 0.8,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7082152974504249,
+      "grad_norm": 0.030330033972859383,
+      "learning_rate": 0.00010915141558577307,
+      "loss": 0.0509,
+      "step": 1250
+    },
+    {
+      "epoch": 0.7082152974504249,
+      "eval_loss": 0.050962403416633606,
+      "eval_runtime": 39.998,
+      "eval_samples_per_second": 25.001,
+      "eval_steps_per_second": 0.8,
+      "step": 1250
+    },
+    {
+      "epoch": 0.7365439093484419,
+      "grad_norm": 0.052799008786678314,
+      "learning_rate": 9.030965963579688e-05,
+      "loss": 0.0583,
+      "step": 1300
+    },
+    {
+      "epoch": 0.7365439093484419,
+      "eval_loss": 0.050432171672582626,
+      "eval_runtime": 39.9904,
+      "eval_samples_per_second": 25.006,
+      "eval_steps_per_second": 0.8,
+      "step": 1300
+    },
+    {
+      "epoch": 0.7648725212464589,
+      "grad_norm": 0.046064119786024094,
+      "learning_rate": 7.288986973708988e-05,
+      "loss": 0.0588,
+      "step": 1350
+    },
+    {
+      "epoch": 0.7648725212464589,
+      "eval_loss": 0.0505264587700367,
+      "eval_runtime": 39.9861,
+      "eval_samples_per_second": 25.009,
+      "eval_steps_per_second": 0.8,
+      "step": 1350
+    },
+    {
+      "epoch": 0.7932011331444759,
+      "grad_norm": 0.1480986326932907,
+      "learning_rate": 5.704716078138403e-05,
+      "loss": 0.045,
+      "step": 1400
+    },
+    {
+      "epoch": 0.7932011331444759,
+      "eval_loss": 0.04980534687638283,
+      "eval_runtime": 39.9962,
+      "eval_samples_per_second": 25.002,
+      "eval_steps_per_second": 0.8,
+      "step": 1400
+    },
+    {
+      "epoch": 0.8215297450424929,
+      "grad_norm": 0.06369854509830475,
+      "learning_rate": 4.2922604509423035e-05,
+      "loss": 0.0473,
+      "step": 1450
+    },
+    {
+      "epoch": 0.8215297450424929,
+      "eval_loss": 0.04981457069516182,
+      "eval_runtime": 39.991,
+      "eval_samples_per_second": 25.006,
+      "eval_steps_per_second": 0.8,
+      "step": 1450
+    },
+    {
+      "epoch": 0.8498583569405099,
+      "grad_norm": 0.09225151687860489,
+      "learning_rate": 3.064197333463017e-05,
+      "loss": 0.0583,
+      "step": 1500
+    },
+    {
+      "epoch": 0.8498583569405099,
+      "eval_loss": 0.04968957602977753,
+      "eval_runtime": 39.993,
+      "eval_samples_per_second": 25.004,
+      "eval_steps_per_second": 0.8,
+      "step": 1500
+    },
+    {
+      "epoch": 0.8781869688385269,
+      "grad_norm": 0.0795036181807518,
+      "learning_rate": 2.0314620399962458e-05,
+      "loss": 0.0466,
+      "step": 1550
+    },
+    {
+      "epoch": 0.8781869688385269,
+      "eval_loss": 0.049295783042907715,
+      "eval_runtime": 40.002,
+      "eval_samples_per_second": 24.999,
+      "eval_steps_per_second": 0.8,
+      "step": 1550
+    },
+    {
+      "epoch": 0.9065155807365439,
+      "grad_norm": 0.1349484771490097,
+      "learning_rate": 1.2032505840510632e-05,
+      "loss": 0.0519,
+      "step": 1600
+    },
+    {
+      "epoch": 0.9065155807365439,
+      "eval_loss": 0.04944948852062225,
+      "eval_runtime": 39.9878,
+      "eval_samples_per_second": 25.008,
+      "eval_steps_per_second": 0.8,
+      "step": 1600
+    },
+    {
+      "epoch": 0.9348441926345609,
+      "grad_norm": 0.14967283606529236,
+      "learning_rate": 5.869377922509161e-06,
+      "loss": 0.0564,
+      "step": 1650
+    },
+    {
+      "epoch": 0.9348441926345609,
+      "eval_loss": 0.04940963163971901,
+      "eval_runtime": 40.0216,
+      "eval_samples_per_second": 24.987,
+      "eval_steps_per_second": 0.8,
+      "step": 1650
+    },
+    {
+      "epoch": 0.9631728045325779,
+      "grad_norm": 0.17137259244918823,
+      "learning_rate": 1.8801163503225082e-06,
+      "loss": 0.049,
+      "step": 1700
+    },
+    {
+      "epoch": 0.9631728045325779,
+      "eval_loss": 0.04926325008273125,
+      "eval_runtime": 40.0111,
+      "eval_samples_per_second": 24.993,
+      "eval_steps_per_second": 0.8,
+      "step": 1700
+    },
+    {
+      "epoch": 0.9915014164305949,
+      "grad_norm": 0.2544344663619995,
+      "learning_rate": 1.0024358894342167e-07,
+      "loss": 0.0496,
+      "step": 1750
+    },
+    {
+      "epoch": 0.9915014164305949,
+      "eval_loss": 0.04924055561423302,
+      "eval_runtime": 40.005,
+      "eval_samples_per_second": 24.997,
+      "eval_steps_per_second": 0.8,
+      "step": 1750
+    },
+    {
+      "epoch": 1.0,
+      "step": 1765,
+      "total_flos": 5.612471411854541e+17,
+      "train_loss": 0.11448637544264537,
+      "train_runtime": 2238.1403,
+      "train_samples_per_second": 6.308,
+      "train_steps_per_second": 0.789
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 1765,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 50.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.612471411854541e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

Llama13B/CMS/t60107d21h14m55,ep=2.0,mlr5.0e-04,b16,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/model_parameters_report.txt

@@ -0,0 +1,409 @@
+=== GLOBAL STATISTICS ===
+Total Parameters: 13,081,728,000
+Trainable Parameters: 65,863,680
+Trainable Percentage: 0.5035%
+
+================================================================================================================================
+=== DETAILED TRAINABLE MATRICES LIST ===
+Layer Name                                                                               | Shape                     | Count          
+---------------------------------------------------------------------------------------------------------------------------------------
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368

Llama13B/CMS/t60107d21h20m27,ep=2.0,mlr5.0e-04,b16,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/model_parameters_report.txt

@@ -0,0 +1,409 @@
+=== GLOBAL STATISTICS ===
+Total Parameters: 13,081,728,000
+Trainable Parameters: 65,863,680
+Trainable Percentage: 0.5035%
+
+================================================================================================================================
+=== DETAILED TRAINABLE MATRICES LIST ===
+Layer Name                                                                               | Shape                     | Count          
+---------------------------------------------------------------------------------------------------------------------------------------
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368

Llama13B/CMS/t60108d00h34m20,ep=2.0,mlr5.0e-04,b8,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/ft/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "base_model_name_or_path": "dfurman/llama-13b",
+  "bias": "none",
+  "col_L": 16,
+  "drop_out": 0.0,
+  "inference_mode": false,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_unique_blocks_L": 32,
+  "num_unique_blocks_R": 16,
+  "peft_type": "SAMA",
+  "revision": null,
+  "row_R": 32,
+  "scaling": 1,
+  "share_factor_L": 1,
+  "share_factor_R": 1,
+  "target_modules": [
+    "up_proj",
+    "down_proj",
+    "q_proj",
+    "k_proj",
+    "v_proj"
+  ],
+  "target_modules_to_skip": null,
+  "task_type": "CAUSAL_LM"
+}

Llama13B/CMS/t60108d00h34m20,ep=2.0,mlr5.0e-04,b8,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/ft/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

Llama13B/CMS/t60108d00h34m20,ep=2.0,mlr5.0e-04,b8,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/ft/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

Llama13B/CMS/t60108d00h34m20,ep=2.0,mlr5.0e-04,b8,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/ft/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "legacy": true,
+  "model_max_length": 512,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

Llama13B/CMS/t60108d00h34m20,ep=2.0,mlr5.0e-04,b8,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/ft2/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "base_model_name_or_path": "dfurman/llama-13b",
+  "bias": "none",
+  "col_L": 16,
+  "drop_out": 0.0,
+  "inference_mode": true,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_unique_blocks_L": 32,
+  "num_unique_blocks_R": 16,
+  "peft_type": "SAMA",
+  "revision": null,
+  "row_R": 32,
+  "scaling": 1,
+  "share_factor_L": 1,
+  "share_factor_R": 1,
+  "target_modules": [
+    "up_proj",
+    "down_proj",
+    "q_proj",
+    "k_proj",
+    "v_proj"
+  ],
+  "target_modules_to_skip": null,
+  "task_type": "CAUSAL_LM"
+}

Llama13B/CMS/t60108d00h34m20,ep=2.0,mlr5.0e-04,b8,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/ft2/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb650adfdb4d1e644d3b081d691f222a3afc5dae46db87118ade70027f6ba1a1
+size 263610019

Llama13B/CMS/t60108d00h34m20,ep=2.0,mlr5.0e-04,b8,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/model_parameters_report.txt

@@ -0,0 +1,409 @@
+=== GLOBAL STATISTICS ===
+Total Parameters: 13,081,728,000
+Trainable Parameters: 65,863,680
+Trainable Percentage: 0.5035%
+
+================================================================================================================================
+=== DETAILED TRAINABLE MATRICES LIST ===
+Layer Name                                                                               | Shape                     | Count          
+---------------------------------------------------------------------------------------------------------------------------------------
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.0.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.0.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.1.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.1.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.2.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.2.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.3.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.3.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.4.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.4.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.5.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.5.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.6.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.6.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.7.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.7.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.8.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.8.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.self_attn.q_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.self_attn.k_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_L.weights      | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.self_attn.v_proj._sama_layer.default.sama_R.weights      | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_L.weights           | [32, 432, 16]             | 221,184
+base_model.model.model.layers.9.mlp.up_proj._sama_layer.default.sama_R.weights           | [16, 32, 320]             | 163,840
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_L.weights         | [32, 160, 16]             | 81,920
+base_model.model.model.layers.9.mlp.down_proj._sama_layer.default.sama_R.weights         | [16, 32, 864]             | 442,368
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.10.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.10.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.11.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.11.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.12.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.12.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.13.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.13.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.14.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.14.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.15.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.15.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.16.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.16.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.17.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.17.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.18.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.18.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.19.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.19.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.20.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.20.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.21.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.21.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.22.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.22.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.23.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.23.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.24.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.24.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.25.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.25.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.26.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.26.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.27.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.27.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.28.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.28.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.29.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.29.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.30.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.30.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.31.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.31.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.32.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.32.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.33.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.33.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.34.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.34.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.35.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.35.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.36.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.36.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.37.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.37.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.38.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.38.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.self_attn.q_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.self_attn.k_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_L.weights     | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.self_attn.v_proj._sama_layer.default.sama_R.weights     | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_L.weights          | [32, 432, 16]             | 221,184
+base_model.model.model.layers.39.mlp.up_proj._sama_layer.default.sama_R.weights          | [16, 32, 320]             | 163,840
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_L.weights        | [32, 160, 16]             | 81,920
+base_model.model.model.layers.39.mlp.down_proj._sama_layer.default.sama_R.weights        | [16, 32, 864]             | 442,368

Llama13B/CMS/t60108d00h34m20,ep=2.0,mlr5.0e-04,b8,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/results/FINAL.json

@@ -0,0 +1,11 @@
+{
+  "boolq": 73.91437308868501,
+  "piqa": 87.97606093579978,
+  "social_i_qa": 82.3439099283521,
+  "hellaswag": 95.4690300736905,
+  "winogrande": 87.05603788476716,
+  "ARC-Easy": 90.65656565656566,
+  "ARC-Challenge": 76.87713310580205,
+  "openbookqa": 89.2,
+  "average_score": 85.4366388342078
+}

Llama13B/CMS/t60108d00h34m20,ep=2.0,mlr5.0e-04,b8,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/trainer_state.json

@@ -0,0 +1,950 @@
+{
+  "best_global_step": 9000,
+  "best_metric": 0.022743234410881996,
+  "best_model_checkpoint": "./Llama13B/CMS/t60108d00h34m20,ep=2.0,mlr5.0e-04,b8,nb32,16,cL16,rR32,s1,initdef,dr0.0,size146627,5/checkpoint-9000",
+  "epoch": 2.0,
+  "eval_steps": 300,
+  "global_step": 18330,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03273501009329478,
+      "grad_norm": 0.0560440868139267,
+      "learning_rate": 0.0004998529736085261,
+      "loss": 0.3967,
+      "step": 300
+    },
+    {
+      "epoch": 0.06547002018658955,
+      "grad_norm": 0.044248245656490326,
+      "learning_rate": 0.0004990760146037058,
+      "loss": 0.0538,
+      "step": 600
+    },
+    {
+      "epoch": 0.06547002018658955,
+      "eval_loss": 0.04312271624803543,
+      "eval_runtime": 42.396,
+      "eval_samples_per_second": 23.587,
+      "eval_steps_per_second": 0.755,
+      "step": 600
+    },
+    {
+      "epoch": 0.09820503027988434,
+      "grad_norm": 0.09440872073173523,
+      "learning_rate": 0.0004976333236993797,
+      "loss": 0.0458,
+      "step": 900
+    },
+    {
+      "epoch": 0.09820503027988434,
+      "eval_loss": 0.03903611749410629,
+      "eval_runtime": 42.1342,
+      "eval_samples_per_second": 23.734,
+      "eval_steps_per_second": 0.759,
+      "step": 900
+    },
+    {
+      "epoch": 0.1309400403731791,
+      "grad_norm": 0.12280289828777313,
+      "learning_rate": 0.0004955287569286458,
+      "loss": 0.0401,
+      "step": 1200
+    },
+    {
+      "epoch": 0.1309400403731791,
+      "eval_loss": 0.03701520711183548,
+      "eval_runtime": 42.1268,
+      "eval_samples_per_second": 23.738,
+      "eval_steps_per_second": 0.76,
+      "step": 1200
+    },
+    {
+      "epoch": 0.1636750504664739,
+      "grad_norm": 0.05425046756863594,
+      "learning_rate": 0.0004927679393904767,
+      "loss": 0.0369,
+      "step": 1500
+    },
+    {
+      "epoch": 0.1636750504664739,
+      "eval_loss": 0.03505582734942436,
+      "eval_runtime": 42.1495,
+      "eval_samples_per_second": 23.725,
+      "eval_steps_per_second": 0.759,
+      "step": 1500
+    },
+    {
+      "epoch": 0.19641006055976867,
+      "grad_norm": 0.062309328466653824,
+      "learning_rate": 0.0004893582502149217,
+      "loss": 0.0361,
+      "step": 1800
+    },
+    {
+      "epoch": 0.19641006055976867,
+      "eval_loss": 0.0326445996761322,
+      "eval_runtime": 42.1829,
+      "eval_samples_per_second": 23.706,
+      "eval_steps_per_second": 0.759,
+      "step": 1800
+    },
+    {
+      "epoch": 0.22914507065306344,
+      "grad_norm": 0.04928302764892578,
+      "learning_rate": 0.0004853088028401219,
+      "loss": 0.0351,
+      "step": 2100
+    },
+    {
+      "epoch": 0.22914507065306344,
+      "eval_loss": 0.03327207267284393,
+      "eval_runtime": 42.1394,
+      "eval_samples_per_second": 23.731,
+      "eval_steps_per_second": 0.759,
+      "step": 2100
+    },
+    {
+      "epoch": 0.2618800807463582,
+      "grad_norm": 0.0912654921412468,
+      "learning_rate": 0.00048063042065385746,
+      "loss": 0.0347,
+      "step": 2400
+    },
+    {
+      "epoch": 0.2618800807463582,
+      "eval_loss": 0.033242832869291306,
+      "eval_runtime": 42.1324,
+      "eval_samples_per_second": 23.735,
+      "eval_steps_per_second": 0.76,
+      "step": 2400
+    },
+    {
+      "epoch": 0.294615090839653,
+      "grad_norm": 0.14929068088531494,
+      "learning_rate": 0.00047533560806473004,
+      "loss": 0.031,
+      "step": 2700
+    },
+    {
+      "epoch": 0.294615090839653,
+      "eval_loss": 0.03134576976299286,
+      "eval_runtime": 42.1366,
+      "eval_samples_per_second": 23.732,
+      "eval_steps_per_second": 0.759,
+      "step": 2700
+    },
+    {
+      "epoch": 0.3273501009329478,
+      "grad_norm": 0.0723583847284317,
+      "learning_rate": 0.0004694385170803028,
+      "loss": 0.0322,
+      "step": 3000
+    },
+    {
+      "epoch": 0.3273501009329478,
+      "eval_loss": 0.03234448283910751,
+      "eval_runtime": 42.1402,
+      "eval_samples_per_second": 23.73,
+      "eval_steps_per_second": 0.759,
+      "step": 3000
+    },
+    {
+      "epoch": 0.3600851110262426,
+      "grad_norm": 0.09747716784477234,
+      "learning_rate": 0.000462954909481528,
+      "loss": 0.0358,
+      "step": 3300
+    },
+    {
+      "epoch": 0.3600851110262426,
+      "eval_loss": 0.02926197461783886,
+      "eval_runtime": 42.1598,
+      "eval_samples_per_second": 23.719,
+      "eval_steps_per_second": 0.759,
+      "step": 3300
+    },
+    {
+      "epoch": 0.39282012111953735,
+      "grad_norm": 0.06375987827777863,
+      "learning_rate": 0.0004559021146945619,
+      "loss": 0.0315,
+      "step": 3600
+    },
+    {
+      "epoch": 0.39282012111953735,
+      "eval_loss": 0.030526984483003616,
+      "eval_runtime": 42.1637,
+      "eval_samples_per_second": 23.717,
+      "eval_steps_per_second": 0.759,
+      "step": 3600
+    },
+    {
+      "epoch": 0.4255551312128321,
+      "grad_norm": 0.037096019834280014,
+      "learning_rate": 0.0004482989834725684,
+      "loss": 0.0329,
+      "step": 3900
+    },
+    {
+      "epoch": 0.4255551312128321,
+      "eval_loss": 0.030683163553476334,
+      "eval_runtime": 42.1644,
+      "eval_samples_per_second": 23.717,
+      "eval_steps_per_second": 0.759,
+      "step": 3900
+    },
+    {
+      "epoch": 0.4582901413061269,
+      "grad_norm": 0.07343820482492447,
+      "learning_rate": 0.0004401658375113103,
+      "loss": 0.032,
+      "step": 4200
+    },
+    {
+      "epoch": 0.4582901413061269,
+      "eval_loss": 0.027738207951188087,
+      "eval_runtime": 42.1656,
+      "eval_samples_per_second": 23.716,
+      "eval_steps_per_second": 0.759,
+      "step": 4200
+    },
+    {
+      "epoch": 0.49102515139942166,
+      "grad_norm": 0.05190209299325943,
+      "learning_rate": 0.0004315244151331965,
+      "loss": 0.0282,
+      "step": 4500
+    },
+    {
+      "epoch": 0.49102515139942166,
+      "eval_loss": 0.02795437164604664,
+      "eval_runtime": 42.1643,
+      "eval_samples_per_second": 23.717,
+      "eval_steps_per_second": 0.759,
+      "step": 4500
+    },
+    {
+      "epoch": 0.5237601614927164,
+      "grad_norm": 0.061979856342077255,
+      "learning_rate": 0.00042239781318496015,
+      "loss": 0.031,
+      "step": 4800
+    },
+    {
+      "epoch": 0.5237601614927164,
+      "eval_loss": 0.028189806267619133,
+      "eval_runtime": 42.1643,
+      "eval_samples_per_second": 23.717,
+      "eval_steps_per_second": 0.759,
+      "step": 4800
+    },
+    {
+      "epoch": 0.5564951715860113,
+      "grad_norm": 0.03137525916099548,
+      "learning_rate": 0.0004128104253042653,
+      "loss": 0.0278,
+      "step": 5100
+    },
+    {
+      "epoch": 0.5564951715860113,
+      "eval_loss": 0.028971396386623383,
+      "eval_runtime": 42.1529,
+      "eval_samples_per_second": 23.723,
+      "eval_steps_per_second": 0.759,
+      "step": 5100
+    },
+    {
+      "epoch": 0.589230181679306,
+      "grad_norm": 0.05158986896276474,
+      "learning_rate": 0.00040278787672024213,
+      "loss": 0.0297,
+      "step": 5400
+    },
+    {
+      "epoch": 0.589230181679306,
+      "eval_loss": 0.027035100385546684,
+      "eval_runtime": 42.1571,
+      "eval_samples_per_second": 23.721,
+      "eval_steps_per_second": 0.759,
+      "step": 5400
+    },
+    {
+      "epoch": 0.6219651917726008,
+      "grad_norm": 0.05266730859875679,
+      "learning_rate": 0.0003923569557622182,
+      "loss": 0.0282,
+      "step": 5700
+    },
+    {
+      "epoch": 0.6219651917726008,
+      "eval_loss": 0.026981640607118607,
+      "eval_runtime": 42.1631,
+      "eval_samples_per_second": 23.717,
+      "eval_steps_per_second": 0.759,
+      "step": 5700
+    },
+    {
+      "epoch": 0.6547002018658956,
+      "grad_norm": 0.05495725944638252,
+      "learning_rate": 0.0003815455422597065,
+      "loss": 0.0301,
+      "step": 6000
+    },
+    {
+      "epoch": 0.6547002018658956,
+      "eval_loss": 0.029135966673493385,
+      "eval_runtime": 42.1443,
+      "eval_samples_per_second": 23.728,
+      "eval_steps_per_second": 0.759,
+      "step": 6000
+    },
+    {
+      "epoch": 0.6874352119591903,
+      "grad_norm": 0.09908170998096466,
+      "learning_rate": 0.00037038253302502694,
+      "loss": 0.0287,
+      "step": 6300
+    },
+    {
+      "epoch": 0.6874352119591903,
+      "eval_loss": 0.027155693620443344,
+      "eval_runtime": 42.1673,
+      "eval_samples_per_second": 23.715,
+      "eval_steps_per_second": 0.759,
+      "step": 6300
+    },
+    {
+      "epoch": 0.7201702220524852,
+      "grad_norm": 0.07481172680854797,
+      "learning_rate": 0.0003588977646177287,
+      "loss": 0.028,
+      "step": 6600
+    },
+    {
+      "epoch": 0.7201702220524852,
+      "eval_loss": 0.02658127434551716,
+      "eval_runtime": 42.157,
+      "eval_samples_per_second": 23.721,
+      "eval_steps_per_second": 0.759,
+      "step": 6600
+    },
+    {
+      "epoch": 0.7529052321457799,
+      "grad_norm": 0.08731873333454132,
+      "learning_rate": 0.000347121933597251,
+      "loss": 0.0278,
+      "step": 6900
+    },
+    {
+      "epoch": 0.7529052321457799,
+      "eval_loss": 0.026023300364613533,
+      "eval_runtime": 42.1259,
+      "eval_samples_per_second": 23.738,
+      "eval_steps_per_second": 0.76,
+      "step": 6900
+    },
+    {
+      "epoch": 0.7856402422390747,
+      "grad_norm": 0.009131520055234432,
+      "learning_rate": 0.0003350865144769708,
+      "loss": 0.0293,
+      "step": 7200
+    },
+    {
+      "epoch": 0.7856402422390747,
+      "eval_loss": 0.02550978772342205,
+      "eval_runtime": 42.1546,
+      "eval_samples_per_second": 23.722,
+      "eval_steps_per_second": 0.759,
+      "step": 7200
+    },
+    {
+      "epoch": 0.8183752523323695,
+      "grad_norm": 0.03787301480770111,
+      "learning_rate": 0.00032282367559893034,
+      "loss": 0.0279,
+      "step": 7500
+    },
+    {
+      "epoch": 0.8183752523323695,
+      "eval_loss": 0.023703226819634438,
+      "eval_runtime": 42.1761,
+      "eval_samples_per_second": 23.71,
+      "eval_steps_per_second": 0.759,
+      "step": 7500
+    },
+    {
+      "epoch": 0.8511102624256642,
+      "grad_norm": 0.05027143657207489,
+      "learning_rate": 0.0003103661931540952,
+      "loss": 0.0265,
+      "step": 7800
+    },
+    {
+      "epoch": 0.8511102624256642,
+      "eval_loss": 0.023319367319345474,
+      "eval_runtime": 42.1616,
+      "eval_samples_per_second": 23.718,
+      "eval_steps_per_second": 0.759,
+      "step": 7800
+    },
+    {
+      "epoch": 0.8838452725189591,
+      "grad_norm": 0.028078876435756683,
+      "learning_rate": 0.0002977473635779503,
+      "loss": 0.028,
+      "step": 8100
+    },
+    {
+      "epoch": 0.8838452725189591,
+      "eval_loss": 0.024727819487452507,
+      "eval_runtime": 42.1437,
+      "eval_samples_per_second": 23.728,
+      "eval_steps_per_second": 0.759,
+      "step": 8100
+    },
+    {
+      "epoch": 0.9165802826122538,
+      "grad_norm": 0.04552329331636429,
+      "learning_rate": 0.0002850009145555825,
+      "loss": 0.0271,
+      "step": 8400
+    },
+    {
+      "epoch": 0.9165802826122538,
+      "eval_loss": 0.023578815162181854,
+      "eval_runtime": 42.1276,
+      "eval_samples_per_second": 23.737,
+      "eval_steps_per_second": 0.76,
+      "step": 8400
+    },
+    {
+      "epoch": 0.9493152927055486,
+      "grad_norm": 0.005148568656295538,
+      "learning_rate": 0.00027216091487411724,
+      "loss": 0.0265,
+      "step": 8700
+    },
+    {
+      "epoch": 0.9493152927055486,
+      "eval_loss": 0.023434242233633995,
+      "eval_runtime": 42.1453,
+      "eval_samples_per_second": 23.727,
+      "eval_steps_per_second": 0.759,
+      "step": 8700
+    },
+    {
+      "epoch": 0.9820503027988433,
+      "grad_norm": 0.05831698328256607,
+      "learning_rate": 0.0002592616833634556,
+      "loss": 0.0242,
+      "step": 9000
+    },
+    {
+      "epoch": 0.9820503027988433,
+      "eval_loss": 0.022743234410881996,
+      "eval_runtime": 42.1372,
+      "eval_samples_per_second": 23.732,
+      "eval_steps_per_second": 0.759,
+      "step": 9000
+    },
+    {
+      "epoch": 1.0147307545419826,
+      "grad_norm": 0.015023964457213879,
+      "learning_rate": 0.00024633769716869496,
+      "loss": 0.0197,
+      "step": 9300
+    },
+    {
+      "epoch": 1.0147307545419826,
+      "eval_loss": 0.02679787389934063,
+      "eval_runtime": 42.1216,
+      "eval_samples_per_second": 23.741,
+      "eval_steps_per_second": 0.76,
+      "step": 9300
+    },
+    {
+      "epoch": 1.0474657646352774,
+      "grad_norm": 0.07885076105594635,
+      "learning_rate": 0.00023342349959940397,
+      "loss": 0.0115,
+      "step": 9600
+    },
+    {
+      "epoch": 1.0474657646352774,
+      "eval_loss": 0.024874603375792503,
+      "eval_runtime": 42.1497,
+      "eval_samples_per_second": 23.725,
+      "eval_steps_per_second": 0.759,
+      "step": 9600
+    },
+    {
+      "epoch": 1.0802007747285722,
+      "grad_norm": 0.12350352108478546,
+      "learning_rate": 0.00022055360780205344,
+      "loss": 0.0105,
+      "step": 9900
+    },
+    {
+      "epoch": 1.0802007747285722,
+      "eval_loss": 0.027364730834960938,
+      "eval_runtime": 42.1561,
+      "eval_samples_per_second": 23.721,
+      "eval_steps_per_second": 0.759,
+      "step": 9900
+    },
+    {
+      "epoch": 1.112935784821867,
+      "grad_norm": 0.18318770825862885,
+      "learning_rate": 0.0002077624205023748,
+      "loss": 0.0117,
+      "step": 10200
+    },
+    {
+      "epoch": 1.112935784821867,
+      "eval_loss": 0.0259011909365654,
+      "eval_runtime": 42.1588,
+      "eval_samples_per_second": 23.72,
+      "eval_steps_per_second": 0.759,
+      "step": 10200
+    },
+    {
+      "epoch": 1.1456707949151617,
+      "grad_norm": 0.10705136507749557,
+      "learning_rate": 0.00019508412606423562,
+      "loss": 0.0117,
+      "step": 10500
+    },
+    {
+      "epoch": 1.1456707949151617,
+      "eval_loss": 0.024886077269911766,
+      "eval_runtime": 42.1435,
+      "eval_samples_per_second": 23.728,
+      "eval_steps_per_second": 0.759,
+      "step": 10500
+    },
+    {
+      "epoch": 1.1784058050084565,
+      "grad_norm": 0.01593521237373352,
+      "learning_rate": 0.00018255261111077092,
+      "loss": 0.0119,
+      "step": 10800
+    },
+    {
+      "epoch": 1.1784058050084565,
+      "eval_loss": 0.02575630508363247,
+      "eval_runtime": 42.1633,
+      "eval_samples_per_second": 23.717,
+      "eval_steps_per_second": 0.759,
+      "step": 10800
+    },
+    {
+      "epoch": 1.2111408151017513,
+      "grad_norm": 0.06906082481145859,
+      "learning_rate": 0.0001702013699520084,
+      "loss": 0.0103,
+      "step": 11100
+    },
+    {
+      "epoch": 1.2111408151017513,
+      "eval_loss": 0.024535585194826126,
+      "eval_runtime": 42.1608,
+      "eval_samples_per_second": 23.719,
+      "eval_steps_per_second": 0.759,
+      "step": 11100
+    },
+    {
+      "epoch": 1.2438758251950461,
+      "grad_norm": 0.0019282581051811576,
+      "learning_rate": 0.00015806341506107184,
+      "loss": 0.0105,
+      "step": 11400
+    },
+    {
+      "epoch": 1.2438758251950461,
+      "eval_loss": 0.023921987041831017,
+      "eval_runtime": 42.1654,
+      "eval_samples_per_second": 23.716,
+      "eval_steps_per_second": 0.759,
+      "step": 11400
+    },
+    {
+      "epoch": 1.276610835288341,
+      "grad_norm": 0.02748137153685093,
+      "learning_rate": 0.00014617118883824025,
+      "loss": 0.0108,
+      "step": 11700
+    },
+    {
+      "epoch": 1.276610835288341,
+      "eval_loss": 0.026060184463858604,
+      "eval_runtime": 42.1556,
+      "eval_samples_per_second": 23.722,
+      "eval_steps_per_second": 0.759,
+      "step": 11700
+    },
+    {
+      "epoch": 1.3093458453816358,
+      "grad_norm": 0.08188096433877945,
+      "learning_rate": 0.00013455647689870125,
+      "loss": 0.0121,
+      "step": 12000
+    },
+    {
+      "epoch": 1.3093458453816358,
+      "eval_loss": 0.025060491636395454,
+      "eval_runtime": 42.1366,
+      "eval_samples_per_second": 23.732,
+      "eval_steps_per_second": 0.759,
+      "step": 12000
+    },
+    {
+      "epoch": 1.3420808554749304,
+      "grad_norm": 0.056115902960300446,
+      "learning_rate": 0.00012325032311576406,
+      "loss": 0.0106,
+      "step": 12300
+    },
+    {
+      "epoch": 1.3420808554749304,
+      "eval_loss": 0.025574443861842155,
+      "eval_runtime": 42.1621,
+      "eval_samples_per_second": 23.718,
+      "eval_steps_per_second": 0.759,
+      "step": 12300
+    },
+    {
+      "epoch": 1.3748158655682252,
+      "grad_norm": 0.012125268578529358,
+      "learning_rate": 0.00011228294664660443,
+      "loss": 0.0099,
+      "step": 12600
+    },
+    {
+      "epoch": 1.3748158655682252,
+      "eval_loss": 0.02508034184575081,
+      "eval_runtime": 42.1596,
+      "eval_samples_per_second": 23.719,
+      "eval_steps_per_second": 0.759,
+      "step": 12600
+    },
+    {
+      "epoch": 1.40755087566152,
+      "grad_norm": 0.0046004848554730415,
+      "learning_rate": 0.00010168366116231597,
+      "loss": 0.0098,
+      "step": 12900
+    },
+    {
+      "epoch": 1.40755087566152,
+      "eval_loss": 0.026252219453454018,
+      "eval_runtime": 42.1483,
+      "eval_samples_per_second": 23.726,
+      "eval_steps_per_second": 0.759,
+      "step": 12900
+    },
+    {
+      "epoch": 1.4402858857548146,
+      "grad_norm": 0.07443796843290329,
+      "learning_rate": 9.148079649815047e-05,
+      "loss": 0.009,
+      "step": 13200
+    },
+    {
+      "epoch": 1.4402858857548146,
+      "eval_loss": 0.026938071474432945,
+      "eval_runtime": 42.1292,
+      "eval_samples_per_second": 23.736,
+      "eval_steps_per_second": 0.76,
+      "step": 13200
+    },
+    {
+      "epoch": 1.4730208958481095,
+      "grad_norm": 0.041219256818294525,
+      "learning_rate": 8.170162293335995e-05,
+      "loss": 0.0095,
+      "step": 13500
+    },
+    {
+      "epoch": 1.4730208958481095,
+      "eval_loss": 0.025707505643367767,
+      "eval_runtime": 42.1816,
+      "eval_samples_per_second": 23.707,
+      "eval_steps_per_second": 0.759,
+      "step": 13500
+    },
+    {
+      "epoch": 1.5057559059414043,
+      "grad_norm": 0.11749689280986786,
+      "learning_rate": 7.237227830302764e-05,
+      "loss": 0.0105,
+      "step": 13800
+    },
+    {
+      "epoch": 1.5057559059414043,
+      "eval_loss": 0.025386406108736992,
+      "eval_runtime": 42.1771,
+      "eval_samples_per_second": 23.71,
+      "eval_steps_per_second": 0.759,
+      "step": 13800
+    },
+    {
+      "epoch": 1.5384909160346991,
+      "grad_norm": 0.06977250427007675,
+      "learning_rate": 6.35176981367018e-05,
+      "loss": 0.01,
+      "step": 14100
+    },
+    {
+      "epoch": 1.5384909160346991,
+      "eval_loss": 0.02462012693285942,
+      "eval_runtime": 42.1422,
+      "eval_samples_per_second": 23.729,
+      "eval_steps_per_second": 0.759,
+      "step": 14100
+    },
+    {
+      "epoch": 1.571225926127994,
+      "grad_norm": 0.022144939750432968,
+      "learning_rate": 5.516154901055917e-05,
+      "loss": 0.0104,
+      "step": 14400
+    },
+    {
+      "epoch": 1.571225926127994,
+      "eval_loss": 0.0242360457777977,
+      "eval_runtime": 42.1513,
+      "eval_samples_per_second": 23.724,
+      "eval_steps_per_second": 0.759,
+      "step": 14400
+    },
+    {
+      "epoch": 1.6039609362212888,
+      "grad_norm": 0.12135706096887589,
+      "learning_rate": 4.732616529123546e-05,
+      "loss": 0.0097,
+      "step": 14700
+    },
+    {
+      "epoch": 1.6039609362212888,
+      "eval_loss": 0.024823835119605064,
+      "eval_runtime": 42.1507,
+      "eval_samples_per_second": 23.724,
+      "eval_steps_per_second": 0.759,
+      "step": 14700
+    },
+    {
+      "epoch": 1.6366959463145836,
+      "grad_norm": 0.011291204020380974,
+      "learning_rate": 4.00324894403929e-05,
+      "loss": 0.0099,
+      "step": 15000
+    },
+    {
+      "epoch": 1.6366959463145836,
+      "eval_loss": 0.025008805096149445,
+      "eval_runtime": 42.1592,
+      "eval_samples_per_second": 23.72,
+      "eval_steps_per_second": 0.759,
+      "step": 15000
+    },
+    {
+      "epoch": 1.6694309564078782,
+      "grad_norm": 0.04522231966257095,
+      "learning_rate": 3.330001603958035e-05,
+      "loss": 0.0103,
+      "step": 15300
+    },
+    {
+      "epoch": 1.6694309564078782,
+      "eval_loss": 0.024476325139403343,
+      "eval_runtime": 42.162,
+      "eval_samples_per_second": 23.718,
+      "eval_steps_per_second": 0.759,
+      "step": 15300
+    },
+    {
+      "epoch": 1.702165966501173,
+      "grad_norm": 0.0071013192646205425,
+      "learning_rate": 2.7146739684996204e-05,
+      "loss": 0.0099,
+      "step": 15600
+    },
+    {
+      "epoch": 1.702165966501173,
+      "eval_loss": 0.02445380762219429,
+      "eval_runtime": 42.1418,
+      "eval_samples_per_second": 23.729,
+      "eval_steps_per_second": 0.759,
+      "step": 15600
+    },
+    {
+      "epoch": 1.7349009765944676,
+      "grad_norm": 0.03472641855478287,
+      "learning_rate": 2.1589106891421295e-05,
+      "loss": 0.0104,
+      "step": 15900
+    },
+    {
+      "epoch": 1.7349009765944676,
+      "eval_loss": 0.024665959179401398,
+      "eval_runtime": 42.1504,
+      "eval_samples_per_second": 23.725,
+      "eval_steps_per_second": 0.759,
+      "step": 15900
+    },
+    {
+      "epoch": 1.7676359866877625,
+      "grad_norm": 0.010884981602430344,
+      "learning_rate": 1.664197213387314e-05,
+      "loss": 0.0098,
+      "step": 16200
+    },
+    {
+      "epoch": 1.7676359866877625,
+      "eval_loss": 0.02443353645503521,
+      "eval_runtime": 42.1323,
+      "eval_samples_per_second": 23.735,
+      "eval_steps_per_second": 0.76,
+      "step": 16200
+    },
+    {
+      "epoch": 1.8003709967810573,
+      "grad_norm": 0.09009554237127304,
+      "learning_rate": 1.2318558144474301e-05,
+      "loss": 0.0099,
+      "step": 16500
+    },
+    {
+      "epoch": 1.8003709967810573,
+      "eval_loss": 0.024443812668323517,
+      "eval_runtime": 42.1454,
+      "eval_samples_per_second": 23.727,
+      "eval_steps_per_second": 0.759,
+      "step": 16500
+    },
+    {
+      "epoch": 1.833106006874352,
+      "grad_norm": 0.0012289569713175297,
+      "learning_rate": 8.630420570652776e-06,
+      "loss": 0.009,
+      "step": 16800
+    },
+    {
+      "epoch": 1.833106006874352,
+      "eval_loss": 0.02449021488428116,
+      "eval_runtime": 42.1851,
+      "eval_samples_per_second": 23.705,
+      "eval_steps_per_second": 0.759,
+      "step": 16800
+    },
+    {
+      "epoch": 1.865841016967647,
+      "grad_norm": 0.007200389634817839,
+      "learning_rate": 5.587417089136487e-06,
+      "loss": 0.0107,
+      "step": 17100
+    },
+    {
+      "epoch": 1.865841016967647,
+      "eval_loss": 0.024273129180073738,
+      "eval_runtime": 42.1539,
+      "eval_samples_per_second": 23.723,
+      "eval_steps_per_second": 0.759,
+      "step": 17100
+    },
+    {
+      "epoch": 1.8985760270609418,
+      "grad_norm": 0.045455873012542725,
+      "learning_rate": 3.197681058294655e-06,
+      "loss": 0.0096,
+      "step": 17400
+    },
+    {
+      "epoch": 1.8985760270609418,
+      "eval_loss": 0.024419987574219704,
+      "eval_runtime": 42.1842,
+      "eval_samples_per_second": 23.706,
+      "eval_steps_per_second": 0.759,
+      "step": 17400
+    },
+    {
+      "epoch": 1.9313110371542366,
+      "grad_norm": 0.12856726348400116,
+      "learning_rate": 1.4675997792476526e-06,
+      "loss": 0.0092,
+      "step": 17700
+    },
+    {
+      "epoch": 1.9313110371542366,
+      "eval_loss": 0.024318819865584373,
+      "eval_runtime": 42.1661,
+      "eval_samples_per_second": 23.716,
+      "eval_steps_per_second": 0.759,
+      "step": 17700
+    },
+    {
+      "epoch": 1.9640460472475312,
+      "grad_norm": 0.045294079929590225,
+      "learning_rate": 4.017974238492983e-07,
+      "loss": 0.0082,
+      "step": 18000
+    },
+    {
+      "epoch": 1.9640460472475312,
+      "eval_loss": 0.024305343627929688,
+      "eval_runtime": 42.1745,
+      "eval_samples_per_second": 23.711,
+      "eval_steps_per_second": 0.759,
+      "step": 18000
+    },
+    {
+      "epoch": 1.996781057340826,
+      "grad_norm": 0.025693634524941444,
+      "learning_rate": 3.122675172106959e-09,
+      "loss": 0.0091,
+      "step": 18300
+    },
+    {
+      "epoch": 1.996781057340826,
+      "eval_loss": 0.02438713051378727,
+      "eval_runtime": 42.1308,
+      "eval_samples_per_second": 23.736,
+      "eval_steps_per_second": 0.76,
+      "step": 18300
+    },
+    {
+      "epoch": 2.0,
+      "step": 18330,
+      "total_flos": 1.1637418990239744e+19,
+      "train_loss": 0.02701456347613535,
+      "train_runtime": 25881.389,
+      "train_samples_per_second": 11.331,
+      "train_steps_per_second": 0.708
+    }
+  ],
+  "logging_steps": 300,
+  "max_steps": 18330,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 300,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.1637418990239744e+19,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

Llama13B/CMS/t60108d07h46m12,ep=2.0,mlr5.0e-04,b8,nb8,8,cL8,rR8,s1,initdef,dr0.0,size146627,5/ft/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "base_model_name_or_path": "dfurman/llama-13b",
+  "bias": "none",
+  "col_L": 8,
+  "drop_out": 0.0,
+  "inference_mode": false,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_unique_blocks_L": 8,
+  "num_unique_blocks_R": 8,
+  "peft_type": "SAMA",
+  "revision": null,
+  "row_R": 8,
+  "scaling": 1,
+  "share_factor_L": 1,
+  "share_factor_R": 1,
+  "target_modules": [
+    "down_proj",
+    "up_proj",
+    "k_proj",
+    "v_proj",
+    "q_proj"
+  ],
+  "target_modules_to_skip": null,
+  "task_type": "CAUSAL_LM"
+}