Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,9 @@ 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
+wandb/run-20241030_011509-cqcwsj7s/run-cqcwsj7s.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20241030_013141-v317zdzd/run-v317zdzd.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20241101_012733-4u8e027p/run-4u8e027p.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20241101_012733-e3zsr634/run-e3zsr634.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20241030_011509-3dp0dtmk/run-3dp0dtmk.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20241129_235322-dmnv987j/run-dmnv987j.wandb filter=lfs diff=lfs merge=lfs -text

wandb/run-20241030_010306-uhzyjdga/files/config.yaml

@@ -0,0 +1,47 @@
+_wandb:
+    value:
+        cli_version: 0.18.5
+        m: []
+        python_version: 3.9.19
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 49
+                - 51
+                - 53
+                - 55
+                - 71
+                - 98
+            "2":
+                - 1
+                - 5
+                - 11
+                - 49
+                - 51
+                - 53
+                - 55
+                - 71
+                - 98
+            "3":
+                - 13
+                - 23
+                - 55
+            "4": 3.9.19
+            "5": 0.18.5
+            "6": 4.45.1
+            "8":
+                - 5
+            "12": 0.18.5
+            "13": linux-x86_64
+batch_size:
+    value: 3
+epoch:
+    value: 7
+perturbation:
+    value: reverse_control
+seed:
+    value: 0
+train_set:
+    value: 10M

wandb/run-20241030_010306-uhzyjdga/files/output.log

@@ -0,0 +1,4 @@
+Traceback (most recent call last):
+  File "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py", line 162, in <module>
+    dataset_name = f"babylm_{args.perturbation}_{args.train_zset}_seed{args.seed}"
+AttributeError: 'Namespace' object has no attribute 'train_zset'

wandb/run-20241030_011509-3dp0dtmk/run-3dp0dtmk.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:774953708a5c0869fb723f0b2ad7b4b7f78c8dee37c2a8c29ac293e0b287dea8
+size 163840

wandb/run-20241030_011509-cqcwsj7s/run-cqcwsj7s.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcc31a33c557e349e62443d6987df97fb3737a5ae668a6b9912256109d629edb
+size 163840

wandb/run-20241030_011509-oh4ul0xu/files/output.log

@@ -0,0 +1,18 @@
+Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:04<00:00,  2.03s/it]
+Map: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 17519/17519 [00:53<00:00, 325.05 examples/s]
+Map: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 18140/18140 [00:55<00:00, 326.84 examples/s]
+tokenized_valid: Dataset({
+    features: ['input_ids', 'attention_mask'],
+    num_rows: 600
+})
+/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/transformers/training_args.py:1545: FutureWarning: `evaluation_strategy` is deprecated and will be removed in version 4.46 of 🤗 Transformers. Use `eval_strategy` instead
+  warnings.warn(
+[2024-10-30 01:17:04,609] [INFO] [real_accelerator.py:219:get_accelerator] Setting ds_accelerator to cuda (auto detect)
+[2024-10-30 01:17:10,763] [INFO] [comm.py:652:init_distributed] cdb=None
+Installed CUDA version 11.8 does not match the version torch was compiled with 11.7 but since the APIs are compatible, accepting this combination
+Using /home/chunhui/.cache/torch_extensions/py39_cu117 as PyTorch extensions root...
+Emitting ninja build file /home/chunhui/.cache/torch_extensions/py39_cu117/cpu_adam/build.ninja...
+Building extension module cpu_adam...
+Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
+Loading extension module cpu_adam...
+Time to load cpu_adam op: 4.44950008392334 seconds

wandb/run-20241030_011509-oh4ul0xu/files/requirements.txt

@@ -0,0 +1,147 @@
+funcsigs==1.0.2
+sentry-sdk==2.17.0
+multiprocess==0.70.16
+numpy==1.26.2
+pluralizer==1.2.0
+debugpy==1.6.7
+nvidia-cudnn-cu11==8.5.0.96
+deepspeed==0.15.2
+data==0.4
+pandas==2.1.3
+tomli==2.0.1
+charset-normalizer==3.3.2
+attrs==24.2.0
+aiosignal==1.3.1
+fsspec==2023.10.0
+nvidia-cusparse-cu11==11.7.4.91
+zipp==3.12.0
+mypy-extensions==1.0.0
+datasets==3.0.1
+joblib==1.3.2
+hjson==3.1.0
+traitlets==5.7.1
+stack-data==0.6.0
+transformers==4.45.1
+sympy==1.11.1
+Pygments==2.15.0
+docker-pycreds==0.4.0
+dill==0.3.8
+wheel==0.44.0
+prompt-toolkit==3.0.30
+parso==0.8.3
+ipykernel==6.23.1
+pyarrow==17.0.0
+certifi==2023.11.17
+nvidia-cufft-cu11==10.9.0.58
+six==1.16.0
+pydantic==2.9.2
+click==8.1.7
+nest-asyncio==1.5.6
+gmpy2==2.1.0
+matplotlib==3.8.2
+scipy==1.11.4
+typing_extensions==4.12.2
+statsmodels==0.14.0
+huggingface-hub==0.25.0
+frozenlist==1.4.1
+gpustat==1.1.1
+nvidia-nvtx-cu11==11.7.91
+safetensors==0.4.5
+stanza==1.9.2
+decorator==5.1.1
+seaborn==0.13.0
+sentencepiece==0.2.0
+PyYAML==6.0.1
+black==24.8.0
+protobuf==4.25.1
+pickleshare==0.7.5
+peft==0.13.0
+triton==2.0.0
+nvidia-cuda-runtime-cu11==11.7.99
+Jinja2==3.1.2
+nvidia-cusolver-cu11==11.4.0.1
+executing==1.2.0
+jupyter_client==8.1.0
+pluggy==1.3.0
+cmake==3.30.3
+pytz==2023.3.post1
+aiohappyeyeballs==2.4.2
+kiwisolver==1.4.5
+py-cpuinfo==9.0.0
+Pillow==10.1.0
+ptyprocess==0.7.0
+importlib_resources==6.4.5
+GitPython==3.1.43
+importlib-metadata==6.0.0
+iniconfig==2.0.0
+scikit-learn==1.3.2
+exceptiongroup==1.1.0
+networkx==2.8.6
+accelerate==1.0.0
+nltk==3.8.1
+shutilwhich==1.1.0
+fonttools==4.45.1
+future==0.18.3
+aiohttp==3.10.6
+wcwidth==0.2.5
+idna==3.6
+filelock==3.12.2
+pathspec==0.12.1
+jupyter_core==5.1.0
+lit==18.1.8
+nvidia-curand-cu11==10.2.10.91
+nvidia-cublas-cu11==11.10.3.66
+nvidia-ml-py==12.560.30
+msgpack==1.1.0
+python-dateutil==2.8.2
+blessed==1.20.0
+packaging==23.0
+gitdb==4.0.11
+yarl==1.13.0
+emoji==2.8.0
+tzdata==2023.3
+cycler==0.12.1
+tornado==6.2
+backcall==0.2.0
+plotnine==0.12.4
+ninja==1.11.1.1
+latex==0.7.0
+wandb==0.18.5
+setproctitle==1.3.3
+threadpoolctl==3.2.0
+requests==2.32.3
+pyparsing==3.1.1
+smmap==5.0.1
+pyzmq==23.0.0
+async-timeout==4.0.3
+annotated-types==0.7.0
+matplotlib-inline==0.1.6
+latexcodec==1.0.0
+ipython==8.0.0
+patsy==0.5.3
+contourpy==1.2.0
+multidict==6.1.0
+mizani==0.9.3
+urllib3==2.1.0
+tokenizers==0.20.0
+MarkupSafe==2.1.2
+pip==24.2
+pexpect==4.8.0
+tqdm==4.66.5
+jedi==0.18.2
+pydantic_core==2.23.4
+tempdir==0.7.1
+mpmath==1.2.1
+setuptools==72.1.0
+pytest==7.4.3
+pure-eval==0.2.2
+psutil==5.9.1
+comm==0.1.2
+nvidia-cuda-cupti-cu11==11.7.101
+nvidia-cuda-nvrtc-cu11==11.7.99
+regex==2023.10.3
+platformdirs==2.5.2
+asttokens==2.2.1
+torch==2.0.0
+nvidia-nccl-cu11==2.14.3
+xxhash==3.5.0

wandb/run-20241030_011509-oh4ul0xu/files/wandb-metadata.json

@@ -0,0 +1,97 @@
+{
+  "os":  "Linux-5.4.0-162-generic-x86_64-with-glibc2.31",
+  "python":  "3.9.19",
+  "startedAt":  "2024-10-30T05:15:09.449635Z",
+  "args":  [
+    "--perturbation",
+    "reverse_control",
+    "--train_set",
+    "10M",
+    "--batch_size",
+    "3",
+    "--epoch",
+    "7",
+    "--seed",
+    "0"
+  ],
+  "program":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py",
+  "codePath":  "train/train_deep_wandb.py",
+  "git":  {
+    "remote":  "git@hf.co:Yaning1001/Impossible_llm.git",
+    "commit":  "ed716cdcfcdea02b67f7ed0f3504c2b1c8b737c4"
+  },
+  "email":  "yaning1001@gmail.com",
+  "root":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train",
+  "host":  "mms-large-2",
+  "username":  "chunhui",
+  "executable":  "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/bin/python",
+  "codePathLocal":  "train_deep_wandb.py",
+  "cpu_count":  32,
+  "cpu_count_logical":  64,
+  "gpu":  "NVIDIA RTX A6000",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1888559353856",
+      "used":  "1719287033856"
+    }
+  },
+  "memory":  {
+    "total":  "202617098240"
+  },
+  "cpu":  {
+    "count":  32,
+    "countLogical":  64
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    }
+  ],
+  "cudaVersion":  "11.8"
+}

wandb/run-20241030_011509-oh4ul0xu/logs/debug-internal.log

@@ -0,0 +1,8 @@
+{"time":"2024-10-30T01:15:09.452194345-04:00","level":"INFO","msg":"using version","core version":"0.18.5"}
+{"time":"2024-10-30T01:15:09.452230726-04:00","level":"INFO","msg":"created symlink","path":"/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241030_011509-oh4ul0xu/logs/debug-core.log"}
+{"time":"2024-10-30T01:15:09.561903697-04:00","level":"INFO","msg":"created new stream","id":"oh4ul0xu"}
+{"time":"2024-10-30T01:15:09.561951377-04:00","level":"INFO","msg":"stream: started","id":"oh4ul0xu"}
+{"time":"2024-10-30T01:15:09.562009358-04:00","level":"INFO","msg":"handler: started","stream_id":{"value":"oh4ul0xu"}}
+{"time":"2024-10-30T01:15:09.562149939-04:00","level":"INFO","msg":"sender: started","stream_id":"oh4ul0xu"}
+{"time":"2024-10-30T01:15:09.561992248-04:00","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"oh4ul0xu"}}
+{"time":"2024-10-30T01:15:09.766220198-04:00","level":"INFO","msg":"Starting system monitor"}

wandb/run-20241030_012617-44bz0olo/logs/debug-internal.log

@@ -0,0 +1,8 @@
+{"time":"2024-10-30T01:26:17.464896093-04:00","level":"INFO","msg":"using version","core version":"0.18.5"}
+{"time":"2024-10-30T01:26:17.464907623-04:00","level":"INFO","msg":"created symlink","path":"/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241030_012617-44bz0olo/logs/debug-core.log"}
+{"time":"2024-10-30T01:26:17.672983424-04:00","level":"INFO","msg":"created new stream","id":"44bz0olo"}
+{"time":"2024-10-30T01:26:17.673056684-04:00","level":"INFO","msg":"stream: started","id":"44bz0olo"}
+{"time":"2024-10-30T01:26:17.673116215-04:00","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"44bz0olo"}}
+{"time":"2024-10-30T01:26:17.673194555-04:00","level":"INFO","msg":"sender: started","stream_id":"44bz0olo"}
+{"time":"2024-10-30T01:26:17.673137025-04:00","level":"INFO","msg":"handler: started","stream_id":{"value":"44bz0olo"}}
+{"time":"2024-10-30T01:26:17.912768248-04:00","level":"INFO","msg":"Starting system monitor"}

wandb/run-20241030_012617-44bz0olo/logs/debug.log

@@ -0,0 +1,26 @@
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_setup.py:_flush():79] Current SDK version is 0.18.5
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_setup.py:_flush():79] Configure stats pid to 332628
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_setup.py:_flush():79] Loading settings from /home/chunhui/.config/wandb/settings
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_setup.py:_flush():79] Loading settings from /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/settings
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_setup.py:_flush():79] Loading settings from environment variables: {}
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': None, '_disable_service': None}
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'train/train_deep_wandb.py', 'program_abspath': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py', 'program': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py'}
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_setup.py:_flush():79] Applying login settings: {}
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_init.py:_log_setup():534] Logging user logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241030_012617-44bz0olo/logs/debug.log
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_init.py:_log_setup():535] Logging internal logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241030_012617-44bz0olo/logs/debug-internal.log
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_init.py:init():621] calling init triggers
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_init.py:init():628] wandb.init called with sweep_config: {}
+config: {}
+2024-10-30 01:26:17,456 INFO    MainThread:332628 [wandb_init.py:init():671] starting backend
+2024-10-30 01:26:17,457 INFO    MainThread:332628 [wandb_init.py:init():675] sending inform_init request
+2024-10-30 01:26:17,457 INFO    MainThread:332628 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2024-10-30 01:26:17,464 INFO    MainThread:332628 [wandb_init.py:init():688] backend started and connected
+2024-10-30 01:26:17,467 INFO    MainThread:332628 [wandb_init.py:init():783] updated telemetry
+2024-10-30 01:26:17,533 INFO    MainThread:332628 [wandb_init.py:init():816] communicating run to backend with 90.0 second timeout
+2024-10-30 01:26:17,904 INFO    MainThread:332628 [wandb_init.py:init():867] starting run threads in backend
+2024-10-30 01:26:18,043 INFO    MainThread:332628 [wandb_run.py:_console_start():2463] atexit reg
+2024-10-30 01:26:18,044 INFO    MainThread:332628 [wandb_run.py:_redirect():2311] redirect: wrap_raw
+2024-10-30 01:26:18,044 INFO    MainThread:332628 [wandb_run.py:_redirect():2376] Wrapping output streams.
+2024-10-30 01:26:18,044 INFO    MainThread:332628 [wandb_run.py:_redirect():2401] Redirects installed.
+2024-10-30 01:26:18,045 INFO    MainThread:332628 [wandb_init.py:init():911] run started, returning control to user process
+2024-10-30 01:26:18,046 INFO    MainThread:332628 [wandb_run.py:_config_callback():1390] config_cb None None {'perturbation': 'reverse_control', 'train_set': '10M', 'batch_size': 3, 'epoch': 7, 'seed': 0}

wandb/run-20241030_012617-591457tl/files/output.log

@@ -0,0 +1,2 @@
+Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:04<00:00,  2.41s/it]
+Map:  11%|██████████████                                                                                                                  | 2000/18140 [00:06<00:53, 303.69 examples/s]

wandb/run-20241030_012617-591457tl/files/requirements.txt

@@ -0,0 +1,147 @@
+funcsigs==1.0.2
+sentry-sdk==2.17.0
+multiprocess==0.70.16
+numpy==1.26.2
+pluralizer==1.2.0
+debugpy==1.6.7
+nvidia-cudnn-cu11==8.5.0.96
+deepspeed==0.15.2
+data==0.4
+pandas==2.1.3
+tomli==2.0.1
+charset-normalizer==3.3.2
+attrs==24.2.0
+aiosignal==1.3.1
+fsspec==2023.10.0
+nvidia-cusparse-cu11==11.7.4.91
+zipp==3.12.0
+mypy-extensions==1.0.0
+datasets==3.0.1
+joblib==1.3.2
+hjson==3.1.0
+traitlets==5.7.1
+stack-data==0.6.0
+transformers==4.45.1
+sympy==1.11.1
+Pygments==2.15.0
+docker-pycreds==0.4.0
+dill==0.3.8
+wheel==0.44.0
+prompt-toolkit==3.0.30
+parso==0.8.3
+ipykernel==6.23.1
+pyarrow==17.0.0
+certifi==2023.11.17
+nvidia-cufft-cu11==10.9.0.58
+six==1.16.0
+pydantic==2.9.2
+click==8.1.7
+nest-asyncio==1.5.6
+gmpy2==2.1.0
+matplotlib==3.8.2
+scipy==1.11.4
+typing_extensions==4.12.2
+statsmodels==0.14.0
+huggingface-hub==0.25.0
+frozenlist==1.4.1
+gpustat==1.1.1
+nvidia-nvtx-cu11==11.7.91
+safetensors==0.4.5
+stanza==1.9.2
+decorator==5.1.1
+seaborn==0.13.0
+sentencepiece==0.2.0
+PyYAML==6.0.1
+black==24.8.0
+protobuf==4.25.1
+pickleshare==0.7.5
+peft==0.13.0
+triton==2.0.0
+nvidia-cuda-runtime-cu11==11.7.99
+Jinja2==3.1.2
+nvidia-cusolver-cu11==11.4.0.1
+executing==1.2.0
+jupyter_client==8.1.0
+pluggy==1.3.0
+cmake==3.30.3
+pytz==2023.3.post1
+aiohappyeyeballs==2.4.2
+kiwisolver==1.4.5
+py-cpuinfo==9.0.0
+Pillow==10.1.0
+ptyprocess==0.7.0
+importlib_resources==6.4.5
+GitPython==3.1.43
+importlib-metadata==6.0.0
+iniconfig==2.0.0
+scikit-learn==1.3.2
+exceptiongroup==1.1.0
+networkx==2.8.6
+accelerate==1.0.0
+nltk==3.8.1
+shutilwhich==1.1.0
+fonttools==4.45.1
+future==0.18.3
+aiohttp==3.10.6
+wcwidth==0.2.5
+idna==3.6
+filelock==3.12.2
+pathspec==0.12.1
+jupyter_core==5.1.0
+lit==18.1.8
+nvidia-curand-cu11==10.2.10.91
+nvidia-cublas-cu11==11.10.3.66
+nvidia-ml-py==12.560.30
+msgpack==1.1.0
+python-dateutil==2.8.2
+blessed==1.20.0
+packaging==23.0
+gitdb==4.0.11
+yarl==1.13.0
+emoji==2.8.0
+tzdata==2023.3
+cycler==0.12.1
+tornado==6.2
+backcall==0.2.0
+plotnine==0.12.4
+ninja==1.11.1.1
+latex==0.7.0
+wandb==0.18.5
+setproctitle==1.3.3
+threadpoolctl==3.2.0
+requests==2.32.3
+pyparsing==3.1.1
+smmap==5.0.1
+pyzmq==23.0.0
+async-timeout==4.0.3
+annotated-types==0.7.0
+matplotlib-inline==0.1.6
+latexcodec==1.0.0
+ipython==8.0.0
+patsy==0.5.3
+contourpy==1.2.0
+multidict==6.1.0
+mizani==0.9.3
+urllib3==2.1.0
+tokenizers==0.20.0
+MarkupSafe==2.1.2
+pip==24.2
+pexpect==4.8.0
+tqdm==4.66.5
+jedi==0.18.2
+pydantic_core==2.23.4
+tempdir==0.7.1
+mpmath==1.2.1
+setuptools==72.1.0
+pytest==7.4.3
+pure-eval==0.2.2
+psutil==5.9.1
+comm==0.1.2
+nvidia-cuda-cupti-cu11==11.7.101
+nvidia-cuda-nvrtc-cu11==11.7.99
+regex==2023.10.3
+platformdirs==2.5.2
+asttokens==2.2.1
+torch==2.0.0
+nvidia-nccl-cu11==2.14.3
+xxhash==3.5.0

wandb/run-20241030_013141-v317zdzd/run-v317zdzd.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7f6e594f9ede19fbbf5de0a7bceb685142e9e3215a6556ae2b238e47e7aba5a
+size 131072

wandb/run-20241030_222932-lsfm0d2q/files/requirements.txt

@@ -0,0 +1,147 @@
+funcsigs==1.0.2
+sentry-sdk==2.17.0
+multiprocess==0.70.16
+numpy==1.26.2
+pluralizer==1.2.0
+debugpy==1.6.7
+nvidia-cudnn-cu11==8.5.0.96
+deepspeed==0.15.2
+data==0.4
+pandas==2.1.3
+tomli==2.0.1
+charset-normalizer==3.3.2
+attrs==24.2.0
+aiosignal==1.3.1
+fsspec==2023.10.0
+nvidia-cusparse-cu11==11.7.4.91
+zipp==3.12.0
+mypy-extensions==1.0.0
+datasets==3.0.1
+joblib==1.3.2
+hjson==3.1.0
+traitlets==5.7.1
+stack-data==0.6.0
+transformers==4.45.1
+sympy==1.11.1
+Pygments==2.15.0
+docker-pycreds==0.4.0
+dill==0.3.8
+wheel==0.44.0
+prompt-toolkit==3.0.30
+parso==0.8.3
+ipykernel==6.23.1
+pyarrow==17.0.0
+certifi==2023.11.17
+nvidia-cufft-cu11==10.9.0.58
+six==1.16.0
+pydantic==2.9.2
+click==8.1.7
+nest-asyncio==1.5.6
+gmpy2==2.1.0
+matplotlib==3.8.2
+scipy==1.11.4
+typing_extensions==4.12.2
+statsmodels==0.14.0
+huggingface-hub==0.25.0
+frozenlist==1.4.1
+gpustat==1.1.1
+nvidia-nvtx-cu11==11.7.91
+safetensors==0.4.5
+stanza==1.9.2
+decorator==5.1.1
+seaborn==0.13.0
+sentencepiece==0.2.0
+PyYAML==6.0.1
+black==24.8.0
+protobuf==4.25.1
+pickleshare==0.7.5
+peft==0.13.0
+triton==2.0.0
+nvidia-cuda-runtime-cu11==11.7.99
+Jinja2==3.1.2
+nvidia-cusolver-cu11==11.4.0.1
+executing==1.2.0
+jupyter_client==8.1.0
+pluggy==1.3.0
+cmake==3.30.3
+pytz==2023.3.post1
+aiohappyeyeballs==2.4.2
+kiwisolver==1.4.5
+py-cpuinfo==9.0.0
+Pillow==10.1.0
+ptyprocess==0.7.0
+importlib_resources==6.4.5
+GitPython==3.1.43
+importlib-metadata==6.0.0
+iniconfig==2.0.0
+scikit-learn==1.3.2
+exceptiongroup==1.1.0
+networkx==2.8.6
+accelerate==1.0.0
+nltk==3.8.1
+shutilwhich==1.1.0
+fonttools==4.45.1
+future==0.18.3
+aiohttp==3.10.6
+wcwidth==0.2.5
+idna==3.6
+filelock==3.12.2
+pathspec==0.12.1
+jupyter_core==5.1.0
+lit==18.1.8
+nvidia-curand-cu11==10.2.10.91
+nvidia-cublas-cu11==11.10.3.66
+nvidia-ml-py==12.560.30
+msgpack==1.1.0
+python-dateutil==2.8.2
+blessed==1.20.0
+packaging==23.0
+gitdb==4.0.11
+yarl==1.13.0
+emoji==2.8.0
+tzdata==2023.3
+cycler==0.12.1
+tornado==6.2
+backcall==0.2.0
+plotnine==0.12.4
+ninja==1.11.1.1
+latex==0.7.0
+wandb==0.18.5
+setproctitle==1.3.3
+threadpoolctl==3.2.0
+requests==2.32.3
+pyparsing==3.1.1
+smmap==5.0.1
+pyzmq==23.0.0
+async-timeout==4.0.3
+annotated-types==0.7.0
+matplotlib-inline==0.1.6
+latexcodec==1.0.0
+ipython==8.0.0
+patsy==0.5.3
+contourpy==1.2.0
+multidict==6.1.0
+mizani==0.9.3
+urllib3==2.1.0
+tokenizers==0.20.0
+MarkupSafe==2.1.2
+pip==24.2
+pexpect==4.8.0
+tqdm==4.66.5
+jedi==0.18.2
+pydantic_core==2.23.4
+tempdir==0.7.1
+mpmath==1.2.1
+setuptools==72.1.0
+pytest==7.4.3
+pure-eval==0.2.2
+psutil==5.9.1
+comm==0.1.2
+nvidia-cuda-cupti-cu11==11.7.101
+nvidia-cuda-nvrtc-cu11==11.7.99
+regex==2023.10.3
+platformdirs==2.5.2
+asttokens==2.2.1
+torch==2.0.0
+nvidia-nccl-cu11==2.14.3
+xxhash==3.5.0

wandb/run-20241030_231835-tt0m4qwl/files/output.log

@@ -0,0 +1,13 @@
+Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:05<00:00,  2.72s/it]
+tokenized_valid: Dataset({
+    features: ['input_ids', 'attention_mask'],
+    num_rows: 600
+})
+/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/transformers/training_args.py:1545: FutureWarning: `evaluation_strategy` is deprecated and will be removed in version 4.46 of 🤗 Transformers. Use `eval_strategy` instead
+  warnings.warn(
+[2024-10-30 23:18:42,992] [INFO] [real_accelerator.py:219:get_accelerator] Setting ds_accelerator to cuda (auto detect)
+[2024-10-30 23:18:52,859] [INFO] [comm.py:652:init_distributed] cdb=None
+Installed CUDA version 11.8 does not match the version torch was compiled with 11.7 but since the APIs are compatible, accepting this combination
+Using /home/chunhui/.cache/torch_extensions/py39_cu117 as PyTorch extensions root...
+Loading extension module cpu_adam...
+Time to load cpu_adam op: 5.861032247543335 seconds

wandb/run-20241030_231835-tt0m4qwl/files/requirements.txt

@@ -0,0 +1,147 @@
+funcsigs==1.0.2
+sentry-sdk==2.17.0
+multiprocess==0.70.16
+numpy==1.26.2
+pluralizer==1.2.0
+debugpy==1.6.7
+nvidia-cudnn-cu11==8.5.0.96
+deepspeed==0.15.2
+data==0.4
+pandas==2.1.3
+tomli==2.0.1
+charset-normalizer==3.3.2
+attrs==24.2.0
+aiosignal==1.3.1
+fsspec==2023.10.0
+nvidia-cusparse-cu11==11.7.4.91
+zipp==3.12.0
+mypy-extensions==1.0.0
+datasets==3.0.1
+joblib==1.3.2
+hjson==3.1.0
+traitlets==5.7.1
+stack-data==0.6.0
+transformers==4.45.1
+sympy==1.11.1
+Pygments==2.15.0
+docker-pycreds==0.4.0
+dill==0.3.8
+wheel==0.44.0
+prompt-toolkit==3.0.30
+parso==0.8.3
+ipykernel==6.23.1
+pyarrow==17.0.0
+certifi==2023.11.17
+nvidia-cufft-cu11==10.9.0.58
+six==1.16.0
+pydantic==2.9.2
+click==8.1.7
+nest-asyncio==1.5.6
+gmpy2==2.1.0
+matplotlib==3.8.2
+scipy==1.11.4
+typing_extensions==4.12.2
+statsmodels==0.14.0
+huggingface-hub==0.25.0
+frozenlist==1.4.1
+gpustat==1.1.1
+nvidia-nvtx-cu11==11.7.91
+safetensors==0.4.5
+stanza==1.9.2
+decorator==5.1.1
+seaborn==0.13.0
+sentencepiece==0.2.0
+PyYAML==6.0.1
+black==24.8.0
+protobuf==4.25.1
+pickleshare==0.7.5
+peft==0.13.0
+triton==2.0.0
+nvidia-cuda-runtime-cu11==11.7.99
+Jinja2==3.1.2
+nvidia-cusolver-cu11==11.4.0.1
+executing==1.2.0
+jupyter_client==8.1.0
+pluggy==1.3.0
+cmake==3.30.3
+pytz==2023.3.post1
+aiohappyeyeballs==2.4.2
+kiwisolver==1.4.5
+py-cpuinfo==9.0.0
+Pillow==10.1.0
+ptyprocess==0.7.0
+importlib_resources==6.4.5
+GitPython==3.1.43
+importlib-metadata==6.0.0
+iniconfig==2.0.0
+scikit-learn==1.3.2
+exceptiongroup==1.1.0
+networkx==2.8.6
+accelerate==1.0.0
+nltk==3.8.1
+shutilwhich==1.1.0
+fonttools==4.45.1
+future==0.18.3
+aiohttp==3.10.6
+wcwidth==0.2.5
+idna==3.6
+filelock==3.12.2
+pathspec==0.12.1
+jupyter_core==5.1.0
+lit==18.1.8
+nvidia-curand-cu11==10.2.10.91
+nvidia-cublas-cu11==11.10.3.66
+nvidia-ml-py==12.560.30
+msgpack==1.1.0
+python-dateutil==2.8.2
+blessed==1.20.0
+packaging==23.0
+gitdb==4.0.11
+yarl==1.13.0
+emoji==2.8.0
+tzdata==2023.3
+cycler==0.12.1
+tornado==6.2
+backcall==0.2.0
+plotnine==0.12.4
+ninja==1.11.1.1
+latex==0.7.0
+wandb==0.18.5
+setproctitle==1.3.3
+threadpoolctl==3.2.0
+requests==2.32.3
+pyparsing==3.1.1
+smmap==5.0.1
+pyzmq==23.0.0
+async-timeout==4.0.3
+annotated-types==0.7.0
+matplotlib-inline==0.1.6
+latexcodec==1.0.0
+ipython==8.0.0
+patsy==0.5.3
+contourpy==1.2.0
+multidict==6.1.0
+mizani==0.9.3
+urllib3==2.1.0
+tokenizers==0.20.0
+MarkupSafe==2.1.2
+pip==24.2
+pexpect==4.8.0
+tqdm==4.66.5
+jedi==0.18.2
+pydantic_core==2.23.4
+tempdir==0.7.1
+mpmath==1.2.1
+setuptools==72.1.0
+pytest==7.4.3
+pure-eval==0.2.2
+psutil==5.9.1
+comm==0.1.2
+nvidia-cuda-cupti-cu11==11.7.101
+nvidia-cuda-nvrtc-cu11==11.7.99
+regex==2023.10.3
+platformdirs==2.5.2
+asttokens==2.2.1
+torch==2.0.0
+nvidia-nccl-cu11==2.14.3
+xxhash==3.5.0

wandb/run-20241031_000839-mehxid7z/files/wandb-metadata.json

@@ -0,0 +1,97 @@
+{
+  "os":  "Linux-5.4.0-162-generic-x86_64-with-glibc2.31",
+  "python":  "3.9.19",
+  "startedAt":  "2024-10-31T04:08:39.206832Z",
+  "args":  [
+    "--perturbation",
+    "reverse_full",
+    "--train_set",
+    "10M",
+    "--batch_size",
+    "3",
+    "--epoch",
+    "6",
+    "--seed",
+    "0"
+  ],
+  "program":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py",
+  "codePath":  "train/train_deep_wandb.py",
+  "git":  {
+    "remote":  "git@hf.co:Yaning1001/Impossible_llm.git",
+    "commit":  "ed716cdcfcdea02b67f7ed0f3504c2b1c8b737c4"
+  },
+  "email":  "yaning1001@gmail.com",
+  "root":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train",
+  "host":  "mms-large-2",
+  "username":  "chunhui",
+  "executable":  "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/bin/python",
+  "codePathLocal":  "train_deep_wandb.py",
+  "cpu_count":  32,
+  "cpu_count_logical":  64,
+  "gpu":  "NVIDIA RTX A6000",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1888559353856",
+      "used":  "1727270539264"
+    }
+  },
+  "memory":  {
+    "total":  "202617098240"
+  },
+  "cpu":  {
+    "count":  32,
+    "countLogical":  64
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    }
+  ],
+  "cudaVersion":  "11.8"
+}

wandb/run-20241031_000839-mehxid7z/logs/debug.log

@@ -0,0 +1,26 @@
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_setup.py:_flush():79] Current SDK version is 0.18.5
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_setup.py:_flush():79] Configure stats pid to 477295
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_setup.py:_flush():79] Loading settings from /home/chunhui/.config/wandb/settings
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_setup.py:_flush():79] Loading settings from /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/settings
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_setup.py:_flush():79] Loading settings from environment variables: {}
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': None, '_disable_service': None}
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'train/train_deep_wandb.py', 'program_abspath': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py', 'program': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py'}
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_setup.py:_flush():79] Applying login settings: {}
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_init.py:_log_setup():534] Logging user logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241031_000839-mehxid7z/logs/debug.log
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_init.py:_log_setup():535] Logging internal logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241031_000839-mehxid7z/logs/debug-internal.log
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_init.py:init():621] calling init triggers
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_init.py:init():628] wandb.init called with sweep_config: {}
+config: {}
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_init.py:init():671] starting backend
+2024-10-31 00:08:39,204 INFO    MainThread:477295 [wandb_init.py:init():675] sending inform_init request
+2024-10-31 00:08:39,206 INFO    MainThread:477295 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2024-10-31 00:08:39,206 INFO    MainThread:477295 [wandb_init.py:init():688] backend started and connected
+2024-10-31 00:08:39,210 INFO    MainThread:477295 [wandb_init.py:init():783] updated telemetry
+2024-10-31 00:08:39,267 INFO    MainThread:477295 [wandb_init.py:init():816] communicating run to backend with 90.0 second timeout
+2024-10-31 00:08:39,494 INFO    MainThread:477295 [wandb_init.py:init():867] starting run threads in backend
+2024-10-31 00:08:39,592 INFO    MainThread:477295 [wandb_run.py:_console_start():2463] atexit reg
+2024-10-31 00:08:39,593 INFO    MainThread:477295 [wandb_run.py:_redirect():2311] redirect: wrap_raw
+2024-10-31 00:08:39,593 INFO    MainThread:477295 [wandb_run.py:_redirect():2376] Wrapping output streams.
+2024-10-31 00:08:39,593 INFO    MainThread:477295 [wandb_run.py:_redirect():2401] Redirects installed.
+2024-10-31 00:08:39,594 INFO    MainThread:477295 [wandb_init.py:init():911] run started, returning control to user process
+2024-10-31 00:08:39,595 INFO    MainThread:477295 [wandb_run.py:_config_callback():1390] config_cb None None {'perturbation': 'reverse_full', 'train_set': '10M', 'batch_size': 3, 'epoch': 6, 'seed': 0, 'lr': 1e-05}

wandb/run-20241031_001055-u3ke9zn3/files/output.log

@@ -0,0 +1,36 @@
+Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:05<00:00,  2.85s/it]
+tokenized_valid: Dataset({
+    features: ['input_ids', 'attention_mask'],
+    num_rows: 600
+})
+/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/transformers/training_args.py:1545: FutureWarning: `evaluation_strategy` is deprecated and will be removed in version 4.46 of 🤗 Transformers. Use `eval_strategy` instead
+  warnings.warn(
+[2024-10-31 00:11:03,867] [INFO] [real_accelerator.py:219:get_accelerator] Setting ds_accelerator to cuda (auto detect)
+[2024-10-31 00:11:12,798] [INFO] [comm.py:652:init_distributed] cdb=None
+Installed CUDA version 11.8 does not match the version torch was compiled with 11.7 but since the APIs are compatible, accepting this combination
+Using /home/chunhui/.cache/torch_extensions/py39_cu117 as PyTorch extensions root...
+Loading extension module cpu_adam...
+Time to load cpu_adam op: 5.4102067947387695 seconds
+Traceback (most recent call last):
+  File "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py", line 220, in <module>
+    trainer.train()
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/transformers/trainer.py", line 2052, in train
+    return inner_training_loop(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/transformers/trainer.py", line 2388, in _inner_training_loop
+    tr_loss_step = self.training_step(model, inputs)
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/transformers/trainer.py", line 3518, in training_step
+    self.accelerator.backward(loss, **kwargs)
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/accelerate/accelerator.py", line 2238, in backward
+    self.deepspeed_engine_wrapped.backward(loss, **kwargs)
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/accelerate/utils/deepspeed.py", line 195, in backward
+    self.engine.step()
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/deepspeed/runtime/engine.py", line 2213, in step
+    self._take_model_step(lr_kwargs)
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/deepspeed/runtime/engine.py", line 2119, in _take_model_step
+    self.optimizer.step()
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/deepspeed/runtime/zero/stage_1_and_2.py", line 1878, in step
+    fp32_partition.to(get_accelerator().current_device_name()).data)
+KeyboardInterrupt
+Error in atexit._run_exitfuncs:
+Traceback (most recent call last):
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/deepspeed/ops/transformer/inference/triton/matmul_ext.py", line 27, in is_nfs_path

wandb/run-20241031_001055-u3ke9zn3/files/requirements.txt

@@ -0,0 +1,147 @@
+funcsigs==1.0.2
+sentry-sdk==2.17.0
+multiprocess==0.70.16
+numpy==1.26.2
+pluralizer==1.2.0
+debugpy==1.6.7
+nvidia-cudnn-cu11==8.5.0.96
+deepspeed==0.15.2
+data==0.4
+pandas==2.1.3
+tomli==2.0.1
+charset-normalizer==3.3.2
+attrs==24.2.0
+aiosignal==1.3.1
+fsspec==2023.10.0
+nvidia-cusparse-cu11==11.7.4.91
+zipp==3.12.0
+mypy-extensions==1.0.0
+datasets==3.0.1
+joblib==1.3.2
+hjson==3.1.0
+traitlets==5.7.1
+stack-data==0.6.0
+transformers==4.45.1
+sympy==1.11.1
+Pygments==2.15.0
+docker-pycreds==0.4.0
+dill==0.3.8
+wheel==0.44.0
+prompt-toolkit==3.0.30
+parso==0.8.3
+ipykernel==6.23.1
+pyarrow==17.0.0
+certifi==2023.11.17
+nvidia-cufft-cu11==10.9.0.58
+six==1.16.0
+pydantic==2.9.2
+click==8.1.7
+nest-asyncio==1.5.6
+gmpy2==2.1.0
+matplotlib==3.8.2
+scipy==1.11.4
+typing_extensions==4.12.2
+statsmodels==0.14.0
+huggingface-hub==0.25.0
+frozenlist==1.4.1
+gpustat==1.1.1
+nvidia-nvtx-cu11==11.7.91
+safetensors==0.4.5
+stanza==1.9.2
+decorator==5.1.1
+seaborn==0.13.0
+sentencepiece==0.2.0
+PyYAML==6.0.1
+black==24.8.0
+protobuf==4.25.1
+pickleshare==0.7.5
+peft==0.13.0
+triton==2.0.0
+nvidia-cuda-runtime-cu11==11.7.99
+Jinja2==3.1.2
+nvidia-cusolver-cu11==11.4.0.1
+executing==1.2.0
+jupyter_client==8.1.0
+pluggy==1.3.0
+cmake==3.30.3
+pytz==2023.3.post1
+aiohappyeyeballs==2.4.2
+kiwisolver==1.4.5
+py-cpuinfo==9.0.0
+Pillow==10.1.0
+ptyprocess==0.7.0
+importlib_resources==6.4.5
+GitPython==3.1.43
+importlib-metadata==6.0.0
+iniconfig==2.0.0
+scikit-learn==1.3.2
+exceptiongroup==1.1.0
+networkx==2.8.6
+accelerate==1.0.0
+nltk==3.8.1
+shutilwhich==1.1.0
+fonttools==4.45.1
+future==0.18.3
+aiohttp==3.10.6
+wcwidth==0.2.5
+idna==3.6
+filelock==3.12.2
+pathspec==0.12.1
+jupyter_core==5.1.0
+lit==18.1.8
+nvidia-curand-cu11==10.2.10.91
+nvidia-cublas-cu11==11.10.3.66
+nvidia-ml-py==12.560.30
+msgpack==1.1.0
+python-dateutil==2.8.2
+blessed==1.20.0
+packaging==23.0
+gitdb==4.0.11
+yarl==1.13.0
+emoji==2.8.0
+tzdata==2023.3
+cycler==0.12.1
+tornado==6.2
+backcall==0.2.0
+plotnine==0.12.4
+ninja==1.11.1.1
+latex==0.7.0
+wandb==0.18.5
+setproctitle==1.3.3
+threadpoolctl==3.2.0
+requests==2.32.3
+pyparsing==3.1.1
+smmap==5.0.1
+pyzmq==23.0.0
+async-timeout==4.0.3
+annotated-types==0.7.0
+matplotlib-inline==0.1.6
+latexcodec==1.0.0
+ipython==8.0.0
+patsy==0.5.3
+contourpy==1.2.0
+multidict==6.1.0
+mizani==0.9.3
+urllib3==2.1.0
+tokenizers==0.20.0
+MarkupSafe==2.1.2
+pip==24.2
+pexpect==4.8.0
+tqdm==4.66.5
+jedi==0.18.2
+pydantic_core==2.23.4
+tempdir==0.7.1
+mpmath==1.2.1
+setuptools==72.1.0
+pytest==7.4.3
+pure-eval==0.2.2
+psutil==5.9.1
+comm==0.1.2
+nvidia-cuda-cupti-cu11==11.7.101
+nvidia-cuda-nvrtc-cu11==11.7.99
+regex==2023.10.3
+platformdirs==2.5.2
+asttokens==2.2.1
+torch==2.0.0
+nvidia-nccl-cu11==2.14.3
+xxhash==3.5.0

wandb/run-20241031_001055-u3ke9zn3/files/wandb-metadata.json

@@ -0,0 +1,97 @@
+{
+  "os":  "Linux-5.4.0-162-generic-x86_64-with-glibc2.31",
+  "python":  "3.9.19",
+  "startedAt":  "2024-10-31T04:10:55.936073Z",
+  "args":  [
+    "--perturbation",
+    "reverse_full",
+    "--train_set",
+    "10M",
+    "--batch_size",
+    "3",
+    "--epoch",
+    "6",
+    "--seed",
+    "0"
+  ],
+  "program":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py",
+  "codePath":  "train/train_deep_wandb.py",
+  "git":  {
+    "remote":  "git@hf.co:Yaning1001/Impossible_llm.git",
+    "commit":  "ed716cdcfcdea02b67f7ed0f3504c2b1c8b737c4"
+  },
+  "email":  "yaning1001@gmail.com",
+  "root":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train",
+  "host":  "mms-large-2",
+  "username":  "chunhui",
+  "executable":  "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/bin/python",
+  "codePathLocal":  "train_deep_wandb.py",
+  "cpu_count":  32,
+  "cpu_count_logical":  64,
+  "gpu":  "NVIDIA RTX A6000",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1888559353856",
+      "used":  "1728856920064"
+    }
+  },
+  "memory":  {
+    "total":  "202617098240"
+  },
+  "cpu":  {
+    "count":  32,
+    "countLogical":  64
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    }
+  ],
+  "cudaVersion":  "11.8"
+}

wandb/run-20241031_001055-u3ke9zn3/logs/debug.log

@@ -0,0 +1,26 @@
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_setup.py:_flush():79] Current SDK version is 0.18.5
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_setup.py:_flush():79] Configure stats pid to 479385
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_setup.py:_flush():79] Loading settings from /home/chunhui/.config/wandb/settings
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_setup.py:_flush():79] Loading settings from /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/settings
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_setup.py:_flush():79] Loading settings from environment variables: {}
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': None, '_disable_service': None}
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'train/train_deep_wandb.py', 'program_abspath': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py', 'program': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py'}
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_setup.py:_flush():79] Applying login settings: {}
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_init.py:_log_setup():534] Logging user logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241031_001055-u3ke9zn3/logs/debug.log
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_init.py:_log_setup():535] Logging internal logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241031_001055-u3ke9zn3/logs/debug-internal.log
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_init.py:init():621] calling init triggers
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_init.py:init():628] wandb.init called with sweep_config: {}
+config: {}
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_init.py:init():671] starting backend
+2024-10-31 00:10:55,934 INFO    MainThread:479385 [wandb_init.py:init():675] sending inform_init request
+2024-10-31 00:10:55,935 INFO    MainThread:479385 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2024-10-31 00:10:55,935 INFO    MainThread:479385 [wandb_init.py:init():688] backend started and connected
+2024-10-31 00:10:55,939 INFO    MainThread:479385 [wandb_init.py:init():783] updated telemetry
+2024-10-31 00:10:55,980 INFO    MainThread:479385 [wandb_init.py:init():816] communicating run to backend with 90.0 second timeout
+2024-10-31 00:10:56,328 INFO    MainThread:479385 [wandb_init.py:init():867] starting run threads in backend
+2024-10-31 00:10:56,445 INFO    MainThread:479385 [wandb_run.py:_console_start():2463] atexit reg
+2024-10-31 00:10:56,445 INFO    MainThread:479385 [wandb_run.py:_redirect():2311] redirect: wrap_raw
+2024-10-31 00:10:56,445 INFO    MainThread:479385 [wandb_run.py:_redirect():2376] Wrapping output streams.
+2024-10-31 00:10:56,445 INFO    MainThread:479385 [wandb_run.py:_redirect():2401] Redirects installed.
+2024-10-31 00:10:56,447 INFO    MainThread:479385 [wandb_init.py:init():911] run started, returning control to user process
+2024-10-31 00:10:56,447 INFO    MainThread:479385 [wandb_run.py:_config_callback():1390] config_cb None None {'perturbation': 'reverse_full', 'train_set': '10M', 'batch_size': 3, 'epoch': 6, 'seed': 0, 'lr': 1e-05}

wandb/run-20241031_122005-xvi7ifef/files/output.log

@@ -0,0 +1,38 @@
+Downloading shards:   0%|                                                                                                                                        | 0/2 [00:22<?, ?it/s]
+Error in sys.excepthook:
+Traceback (most recent call last):
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/wandb/sdk/lib/exit_hooks.py", line 52, in exc_handler
+    traceback.print_exception(exc_type, exc, tb)
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/traceback.py", line 103, in print_exception
+    for line in TracebackException(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/traceback.py", line 475, in __init__
+    _seen = set()
+KeyboardInterrupt
+
+Original exception was:
+Traceback (most recent call last):
+  File "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py", line 173, in <module>
+    model = AutoModelForCausalLM.from_pretrained(model_name,
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/transformers/models/auto/auto_factory.py", line 564, in from_pretrained
+    return model_class.from_pretrained(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/transformers/modeling_utils.py", line 3769, in from_pretrained
+    resolved_archive_file, sharded_metadata = get_checkpoint_shard_files(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/transformers/utils/hub.py", line 1098, in get_checkpoint_shard_files
+    cached_filename = cached_file(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/transformers/utils/hub.py", line 403, in cached_file
+    resolved_file = hf_hub_download(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/huggingface_hub/utils/_deprecation.py", line 101, in inner_f
+    return f(*args, **kwargs)
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/huggingface_hub/utils/_validators.py", line 114, in _inner_fn
+    return fn(*args, **kwargs)
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/huggingface_hub/file_download.py", line 1232, in hf_hub_download
+    return _hf_hub_download_to_cache_dir(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/huggingface_hub/file_download.py", line 1380, in _hf_hub_download_to_cache_dir
+    with WeakFileLock(lock_path):
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/contextlib.py", line 119, in __enter__
+    return next(self.gen)
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/huggingface_hub/utils/_fixes.py", line 98, in WeakFileLock
+    lock.acquire()
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/filelock/_api.py", line 225, in acquire
+    time.sleep(poll_interval)
+KeyboardInterrupt

wandb/run-20241031_122005-xvi7ifef/files/wandb-metadata.json

@@ -0,0 +1,97 @@
+{
+  "os":  "Linux-5.4.0-162-generic-x86_64-with-glibc2.31",
+  "python":  "3.9.19",
+  "startedAt":  "2024-10-31T16:20:05.891835Z",
+  "args":  [
+    "--perturbation",
+    "reverse_full",
+    "--train_set",
+    "10M",
+    "--batch_size",
+    "3",
+    "--epoch",
+    "6",
+    "--seed",
+    "0"
+  ],
+  "program":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py",
+  "codePath":  "train/train_deep_wandb.py",
+  "git":  {
+    "remote":  "git@hf.co:Yaning1001/Impossible_llm.git",
+    "commit":  "ed716cdcfcdea02b67f7ed0f3504c2b1c8b737c4"
+  },
+  "email":  "yaning1001@gmail.com",
+  "root":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train",
+  "host":  "mms-large-2",
+  "username":  "chunhui",
+  "executable":  "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/bin/python",
+  "codePathLocal":  "train_deep_wandb.py",
+  "cpu_count":  32,
+  "cpu_count_logical":  64,
+  "gpu":  "NVIDIA RTX A6000",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1888559353856",
+      "used":  "1753159847936"
+    }
+  },
+  "memory":  {
+    "total":  "202617098240"
+  },
+  "cpu":  {
+    "count":  32,
+    "countLogical":  64
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    }
+  ],
+  "cudaVersion":  "11.8"
+}

wandb/run-20241031_122005-xvi7ifef/files/wandb-summary.json

@@ -0,0 +1 @@
+{"_wandb":{"runtime":23}}

wandb/run-20241031_122005-xvi7ifef/logs/debug-internal.log

@@ -0,0 +1,11 @@
+{"time":"2024-10-31T12:20:05.893701815-04:00","level":"INFO","msg":"using version","core version":"0.18.5"}
+{"time":"2024-10-31T12:20:05.893713785-04:00","level":"INFO","msg":"created symlink","path":"/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241031_122005-xvi7ifef/logs/debug-core.log"}
+{"time":"2024-10-31T12:20:06.001382077-04:00","level":"INFO","msg":"created new stream","id":"xvi7ifef"}
+{"time":"2024-10-31T12:20:06.001427307-04:00","level":"INFO","msg":"stream: started","id":"xvi7ifef"}
+{"time":"2024-10-31T12:20:06.001462087-04:00","level":"INFO","msg":"sender: started","stream_id":"xvi7ifef"}
+{"time":"2024-10-31T12:20:06.001438197-04:00","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"xvi7ifef"}}
+{"time":"2024-10-31T12:20:06.001453797-04:00","level":"INFO","msg":"handler: started","stream_id":{"value":"xvi7ifef"}}
+{"time":"2024-10-31T12:20:06.189578398-04:00","level":"INFO","msg":"Starting system monitor"}
+{"time":"2024-10-31T12:20:29.345121727-04:00","level":"INFO","msg":"stream: closing","id":"xvi7ifef"}
+{"time":"2024-10-31T12:20:29.345240347-04:00","level":"INFO","msg":"Stopping system monitor"}
+{"time":"2024-10-31T12:20:29.346116817-04:00","level":"INFO","msg":"Stopped system monitor"}

wandb/run-20241031_122005-xvi7ifef/logs/debug.log

@@ -0,0 +1,27 @@
+2024-10-31 12:20:05,888 INFO    MainThread:557187 [wandb_setup.py:_flush():79] Current SDK version is 0.18.5
+2024-10-31 12:20:05,888 INFO    MainThread:557187 [wandb_setup.py:_flush():79] Configure stats pid to 557187
+2024-10-31 12:20:05,888 INFO    MainThread:557187 [wandb_setup.py:_flush():79] Loading settings from /home/chunhui/.config/wandb/settings
+2024-10-31 12:20:05,888 INFO    MainThread:557187 [wandb_setup.py:_flush():79] Loading settings from /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/settings
+2024-10-31 12:20:05,888 INFO    MainThread:557187 [wandb_setup.py:_flush():79] Loading settings from environment variables: {}
+2024-10-31 12:20:05,888 INFO    MainThread:557187 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': None, '_disable_service': None}
+2024-10-31 12:20:05,888 INFO    MainThread:557187 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'train/train_deep_wandb.py', 'program_abspath': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py', 'program': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py'}
+2024-10-31 12:20:05,889 INFO    MainThread:557187 [wandb_setup.py:_flush():79] Applying login settings: {}
+2024-10-31 12:20:05,889 INFO    MainThread:557187 [wandb_init.py:_log_setup():534] Logging user logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241031_122005-xvi7ifef/logs/debug.log
+2024-10-31 12:20:05,889 INFO    MainThread:557187 [wandb_init.py:_log_setup():535] Logging internal logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241031_122005-xvi7ifef/logs/debug-internal.log
+2024-10-31 12:20:05,889 INFO    MainThread:557187 [wandb_init.py:init():621] calling init triggers
+2024-10-31 12:20:05,889 INFO    MainThread:557187 [wandb_init.py:init():628] wandb.init called with sweep_config: {}
+config: {}
+2024-10-31 12:20:05,889 INFO    MainThread:557187 [wandb_init.py:init():671] starting backend
+2024-10-31 12:20:05,889 INFO    MainThread:557187 [wandb_init.py:init():675] sending inform_init request
+2024-10-31 12:20:05,891 INFO    MainThread:557187 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2024-10-31 12:20:05,891 INFO    MainThread:557187 [wandb_init.py:init():688] backend started and connected
+2024-10-31 12:20:05,894 INFO    MainThread:557187 [wandb_init.py:init():783] updated telemetry
+2024-10-31 12:20:05,922 INFO    MainThread:557187 [wandb_init.py:init():816] communicating run to backend with 90.0 second timeout
+2024-10-31 12:20:06,186 INFO    MainThread:557187 [wandb_init.py:init():867] starting run threads in backend
+2024-10-31 12:20:06,278 INFO    MainThread:557187 [wandb_run.py:_console_start():2463] atexit reg
+2024-10-31 12:20:06,278 INFO    MainThread:557187 [wandb_run.py:_redirect():2311] redirect: wrap_raw
+2024-10-31 12:20:06,278 INFO    MainThread:557187 [wandb_run.py:_redirect():2376] Wrapping output streams.
+2024-10-31 12:20:06,278 INFO    MainThread:557187 [wandb_run.py:_redirect():2401] Redirects installed.
+2024-10-31 12:20:06,279 INFO    MainThread:557187 [wandb_init.py:init():911] run started, returning control to user process
+2024-10-31 12:20:06,280 INFO    MainThread:557187 [wandb_run.py:_config_callback():1390] config_cb None None {'perturbation': 'reverse_full', 'train_set': '10M', 'batch_size': 3, 'epoch': 6, 'seed': 0, 'lr': 5e-06}
+2024-10-31 12:20:29,345 WARNING MsgRouterThr:557187 [router.py:message_loop():77] message_loop has been closed

wandb/run-20241101_012438-tr39r2kv/files/wandb-metadata.json

@@ -0,0 +1,29 @@
+{
+  "os":  "Linux-5.4.0-162-generic-x86_64-with-glibc2.31",
+  "python":  "3.9.19",
+  "startedAt":  "2024-11-01T05:24:38.166620Z",
+  "args":  [
+    "--perturbation",
+    "shuffle_nodeterministic",
+    "--train_set",
+    "10M",
+    "--batch_size",
+    "3",
+    "--epoch",
+    "6",
+    "--seed",
+    "0"
+  ],
+  "program":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py",
+  "codePath":  "train/train_deep_wandb.py",
+  "git":  {
+    "remote":  "git@hf.co:Yaning1001/Impossible_llm.git",
+    "commit":  "ed716cdcfcdea02b67f7ed0f3504c2b1c8b737c4"
+  },
+  "email":  "yaning1001@gmail.com",
+  "root":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train",
+  "host":  "mms-large-2",
+  "username":  "chunhui",
+  "executable":  "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/bin/python",
+  "codePathLocal":  "train_deep_wandb.py"
+}

wandb/run-20241101_012438-tr39r2kv/logs/debug.log

@@ -0,0 +1,27 @@
+2024-11-01 01:24:38,164 INFO    MainThread:676350 [wandb_setup.py:_flush():79] Current SDK version is 0.18.5
+2024-11-01 01:24:38,164 INFO    MainThread:676350 [wandb_setup.py:_flush():79] Configure stats pid to 676350
+2024-11-01 01:24:38,164 INFO    MainThread:676350 [wandb_setup.py:_flush():79] Loading settings from /home/chunhui/.config/wandb/settings
+2024-11-01 01:24:38,164 INFO    MainThread:676350 [wandb_setup.py:_flush():79] Loading settings from /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/settings
+2024-11-01 01:24:38,164 INFO    MainThread:676350 [wandb_setup.py:_flush():79] Loading settings from environment variables: {}
+2024-11-01 01:24:38,165 INFO    MainThread:676350 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': None, '_disable_service': None}
+2024-11-01 01:24:38,165 INFO    MainThread:676350 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'train/train_deep_wandb.py', 'program_abspath': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py', 'program': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py'}
+2024-11-01 01:24:38,165 INFO    MainThread:676350 [wandb_setup.py:_flush():79] Applying login settings: {}
+2024-11-01 01:24:38,165 INFO    MainThread:676350 [wandb_init.py:_log_setup():534] Logging user logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241101_012438-tr39r2kv/logs/debug.log
+2024-11-01 01:24:38,165 INFO    MainThread:676350 [wandb_init.py:_log_setup():535] Logging internal logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241101_012438-tr39r2kv/logs/debug-internal.log
+2024-11-01 01:24:38,165 INFO    MainThread:676350 [wandb_init.py:init():621] calling init triggers
+2024-11-01 01:24:38,165 INFO    MainThread:676350 [wandb_init.py:init():628] wandb.init called with sweep_config: {}
+config: {}
+2024-11-01 01:24:38,165 INFO    MainThread:676350 [wandb_init.py:init():671] starting backend
+2024-11-01 01:24:38,165 INFO    MainThread:676350 [wandb_init.py:init():675] sending inform_init request
+2024-11-01 01:24:38,166 INFO    MainThread:676350 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2024-11-01 01:24:38,166 INFO    MainThread:676350 [wandb_init.py:init():688] backend started and connected
+2024-11-01 01:24:38,169 INFO    MainThread:676350 [wandb_init.py:init():783] updated telemetry
+2024-11-01 01:24:38,201 INFO    MainThread:676350 [wandb_init.py:init():816] communicating run to backend with 90.0 second timeout
+2024-11-01 01:24:38,462 INFO    MainThread:676350 [wandb_init.py:init():867] starting run threads in backend
+2024-11-01 01:24:38,576 INFO    MainThread:676350 [wandb_run.py:_console_start():2463] atexit reg
+2024-11-01 01:24:38,577 INFO    MainThread:676350 [wandb_run.py:_redirect():2311] redirect: wrap_raw
+2024-11-01 01:24:38,577 INFO    MainThread:676350 [wandb_run.py:_redirect():2376] Wrapping output streams.
+2024-11-01 01:24:38,577 INFO    MainThread:676350 [wandb_run.py:_redirect():2401] Redirects installed.
+2024-11-01 01:24:38,579 INFO    MainThread:676350 [wandb_init.py:init():911] run started, returning control to user process
+2024-11-01 01:24:38,579 INFO    MainThread:676350 [wandb_run.py:_config_callback():1390] config_cb None None {'perturbation': 'shuffle_nodeterministic', 'train_set': '10M', 'batch_size': 3, 'epoch': 6, 'seed': 0, 'lr': 5e-06}
+2024-11-01 01:24:38,591 WARNING MsgRouterThr:676350 [router.py:message_loop():77] message_loop has been closed

wandb/run-20241101_012612-2icf8uyp/files/output.log

@@ -0,0 +1,12 @@
+Traceback (most recent call last):
+  File "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py", line 164, in <module>
+    dataset = load_dataset('babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/datasets/load.py", line 2074, in load_dataset
+    builder_instance = load_dataset_builder(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/datasets/load.py", line 1832, in load_dataset_builder
+    builder_instance: DatasetBuilder = builder_cls(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/datasets/builder.py", line 342, in __init__
+    self.config, self.config_id = self._create_builder_config(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/datasets/builder.py", line 569, in _create_builder_config
+    raise ValueError(
+ValueError: BuilderConfig 'babylm_shuffle_nodeterministic_10M_seed0' not found. Available: ['babylm_hop_control_10M_seed0', 'babylm_hop_tokens4_10M_seed0', 'babylm_hop_words4_10M_seed0', 'babylm_reverse_control_10M_seed0', 'babylm_reverse_partial_10M_seed0', 'babylm_reverse_full_10M_seed0', 'babylm_shuffle_control_10M_seed0', 'babylm_shuffle_nondeterministic_10M_seed0', 'babylm_shuffle_deterministic21_10M_seed0', 'babylm_shuffle_deterministic57_10M_seed0', 'babylm_shuffle_deterministic84_10M_seed0', 'babylm_shuffle_local3_10M_seed0', 'babylm_shuffle_local5_10M_seed0', 'babylm_shuffle_local10_10M_seed0', 'babylm_shuffle_even_odd_10M_seed0']

wandb/run-20241101_012612-2icf8uyp/files/wandb-summary.json

@@ -0,0 +1 @@
+{"_wandb":{"runtime":0}}

wandb/run-20241101_012612-2icf8uyp/logs/debug-internal.log

@@ -0,0 +1,16 @@
+{"time":"2024-11-01T01:26:12.959111943-04:00","level":"INFO","msg":"using version","core version":"0.18.5"}
+{"time":"2024-11-01T01:26:12.959129113-04:00","level":"INFO","msg":"created symlink","path":"/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241101_012612-2icf8uyp/logs/debug-core.log"}
+{"time":"2024-11-01T01:26:13.06672005-04:00","level":"INFO","msg":"created new stream","id":"2icf8uyp"}
+{"time":"2024-11-01T01:26:13.06676589-04:00","level":"INFO","msg":"stream: started","id":"2icf8uyp"}
+{"time":"2024-11-01T01:26:13.067468275-04:00","level":"INFO","msg":"sender: started","stream_id":"2icf8uyp"}
+{"time":"2024-11-01T01:26:13.06678695-04:00","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"2icf8uyp"}}
+{"time":"2024-11-01T01:26:13.0668113-04:00","level":"INFO","msg":"handler: started","stream_id":{"value":"2icf8uyp"}}
+{"time":"2024-11-01T01:26:13.261986126-04:00","level":"INFO","msg":"Starting system monitor"}
+{"time":"2024-11-01T01:26:13.37198602-04:00","level":"INFO","msg":"stream: closing","id":"2icf8uyp"}
+{"time":"2024-11-01T01:26:13.37204529-04:00","level":"INFO","msg":"Stopping system monitor"}
+{"time":"2024-11-01T01:26:13.372541984-04:00","level":"INFO","msg":"Stopped system monitor"}
+{"time":"2024-11-01T01:26:14.110429588-04:00","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2024-11-01T01:26:14.235735809-04:00","level":"INFO","msg":"handler: closed","stream_id":{"value":"2icf8uyp"}}
+{"time":"2024-11-01T01:26:14.235805009-04:00","level":"INFO","msg":"sender: closed","stream_id":"2icf8uyp"}
+{"time":"2024-11-01T01:26:14.235790089-04:00","level":"INFO","msg":"writer: Close: closed","stream_id":{"value":"2icf8uyp"}}
+{"time":"2024-11-01T01:26:14.236052171-04:00","level":"INFO","msg":"stream: closed","id":"2icf8uyp"}

wandb/run-20241101_012612-kwpxhz4q/files/output.log

@@ -0,0 +1,12 @@
+Traceback (most recent call last):
+  File "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py", line 164, in <module>
+    dataset = load_dataset('babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/datasets/load.py", line 2074, in load_dataset
+    builder_instance = load_dataset_builder(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/datasets/load.py", line 1832, in load_dataset_builder
+    builder_instance: DatasetBuilder = builder_cls(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/datasets/builder.py", line 342, in __init__
+    self.config, self.config_id = self._create_builder_config(
+  File "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/datasets/builder.py", line 569, in _create_builder_config
+    raise ValueError(
+ValueError: BuilderConfig 'babylm_shuffle_nodeterministic_10M_seed0' not found. Available: ['babylm_hop_control_10M_seed0', 'babylm_hop_tokens4_10M_seed0', 'babylm_hop_words4_10M_seed0', 'babylm_reverse_control_10M_seed0', 'babylm_reverse_partial_10M_seed0', 'babylm_reverse_full_10M_seed0', 'babylm_shuffle_control_10M_seed0', 'babylm_shuffle_nondeterministic_10M_seed0', 'babylm_shuffle_deterministic21_10M_seed0', 'babylm_shuffle_deterministic57_10M_seed0', 'babylm_shuffle_deterministic84_10M_seed0', 'babylm_shuffle_local3_10M_seed0', 'babylm_shuffle_local5_10M_seed0', 'babylm_shuffle_local10_10M_seed0', 'babylm_shuffle_even_odd_10M_seed0']

wandb/run-20241101_012612-kwpxhz4q/logs/debug.log

@@ -0,0 +1,27 @@
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_setup.py:_flush():79] Current SDK version is 0.18.5
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_setup.py:_flush():79] Configure stats pid to 677638
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_setup.py:_flush():79] Loading settings from /home/chunhui/.config/wandb/settings
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_setup.py:_flush():79] Loading settings from /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/settings
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_setup.py:_flush():79] Loading settings from environment variables: {}
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': None, '_disable_service': None}
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'train/train_deep_wandb.py', 'program_abspath': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py', 'program': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py'}
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_setup.py:_flush():79] Applying login settings: {}
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_init.py:_log_setup():534] Logging user logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241101_012612-kwpxhz4q/logs/debug.log
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_init.py:_log_setup():535] Logging internal logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241101_012612-kwpxhz4q/logs/debug-internal.log
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_init.py:init():621] calling init triggers
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_init.py:init():628] wandb.init called with sweep_config: {}
+config: {}
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_init.py:init():671] starting backend
+2024-11-01 01:26:12,562 INFO    MainThread:677638 [wandb_init.py:init():675] sending inform_init request
+2024-11-01 01:26:12,563 INFO    MainThread:677638 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2024-11-01 01:26:12,564 INFO    MainThread:677638 [wandb_init.py:init():688] backend started and connected
+2024-11-01 01:26:12,568 INFO    MainThread:677638 [wandb_init.py:init():783] updated telemetry
+2024-11-01 01:26:12,593 INFO    MainThread:677638 [wandb_init.py:init():816] communicating run to backend with 90.0 second timeout
+2024-11-01 01:26:12,862 INFO    MainThread:677638 [wandb_init.py:init():867] starting run threads in backend
+2024-11-01 01:26:12,965 INFO    MainThread:677638 [wandb_run.py:_console_start():2463] atexit reg
+2024-11-01 01:26:12,965 INFO    MainThread:677638 [wandb_run.py:_redirect():2311] redirect: wrap_raw
+2024-11-01 01:26:12,965 INFO    MainThread:677638 [wandb_run.py:_redirect():2376] Wrapping output streams.
+2024-11-01 01:26:12,965 INFO    MainThread:677638 [wandb_run.py:_redirect():2401] Redirects installed.
+2024-11-01 01:26:12,967 INFO    MainThread:677638 [wandb_init.py:init():911] run started, returning control to user process
+2024-11-01 01:26:12,967 INFO    MainThread:677638 [wandb_run.py:_config_callback():1390] config_cb None None {'perturbation': 'shuffle_nodeterministic', 'train_set': '10M', 'batch_size': 3, 'epoch': 6, 'seed': 0, 'lr': 5e-06}
+2024-11-01 01:26:12,976 WARNING MsgRouterThr:677638 [router.py:message_loop():77] message_loop has been closed

wandb/run-20241101_012733-4u8e027p/run-4u8e027p.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8186a12a300f9a4c63bfd615e8ba5e7dfb9ff87f03aa3b2e0db9927c1fbd2c93
+size 1015808

wandb/run-20241101_012733-e3zsr634/run-e3zsr634.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94ee42471a30396c5c6bc1d3008739f6570cb77323089d6b7050c5abd4de1101
+size 1015808

wandb/run-20241101_012734-m18lsdzn/files/output.log

@@ -0,0 +1,19 @@
+model.safetensors.index.json: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20.9k/20.9k [00:00<00:00, 4.24MB/s]
+model-00001-of-00002.safetensors: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████| 4.97G/4.97G [01:58<00:00, 42.0MB/s]
+model-00002-of-00002.safetensors: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████| 1.46G/1.46G [00:34<00:00, 42.5MB/s]
+Downloading shards: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [02:32<00:00, 76.36s/it]
+Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:06<00:00,  3.43s/it]
+Map: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 16425/16425 [00:52<00:00, 314.14 examples/s]
+Map: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 17013/17013 [00:53<00:00, 317.54 examples/s]
+tokenized_valid: Dataset({
+    features: ['input_ids', 'attention_mask'],
+    num_rows: 600
+})
+/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/lib/python3.9/site-packages/transformers/training_args.py:1545: FutureWarning: `evaluation_strategy` is deprecated and will be removed in version 4.46 of 🤗 Transformers. Use `eval_strategy` instead
+  warnings.warn(
+[2024-11-01 01:32:29,634] [INFO] [real_accelerator.py:219:get_accelerator] Setting ds_accelerator to cuda (auto detect)
+[2024-11-01 01:32:39,928] [INFO] [comm.py:652:init_distributed] cdb=None
+Installed CUDA version 11.8 does not match the version torch was compiled with 11.7 but since the APIs are compatible, accepting this combination
+Using /home/chunhui/.cache/torch_extensions/py39_cu117 as PyTorch extensions root...
+Loading extension module cpu_adam...
+Time to load cpu_adam op: 5.3903725147247314 seconds

wandb/run-20241101_200502-8iosiep6/files/output.log

@@ -0,0 +1 @@
+Loading checkpoint shards:   0%|                                                                                                                                                        | 0/2 [00:00<?, ?it/s]

wandb/run-20241101_200502-8iosiep6/files/requirements.txt

@@ -0,0 +1,147 @@
+funcsigs==1.0.2
+sentry-sdk==2.17.0
+multiprocess==0.70.16
+numpy==1.26.2
+pluralizer==1.2.0
+debugpy==1.6.7
+nvidia-cudnn-cu11==8.5.0.96
+deepspeed==0.15.2
+data==0.4
+pandas==2.1.3
+tomli==2.0.1
+charset-normalizer==3.3.2
+attrs==24.2.0
+aiosignal==1.3.1
+fsspec==2023.10.0
+nvidia-cusparse-cu11==11.7.4.91
+zipp==3.12.0
+mypy-extensions==1.0.0
+datasets==3.0.1
+joblib==1.3.2
+hjson==3.1.0
+traitlets==5.7.1
+stack-data==0.6.0
+transformers==4.45.1
+sympy==1.11.1
+Pygments==2.15.0
+docker-pycreds==0.4.0
+dill==0.3.8
+wheel==0.44.0
+prompt-toolkit==3.0.30
+parso==0.8.3
+ipykernel==6.23.1
+pyarrow==17.0.0
+certifi==2023.11.17
+nvidia-cufft-cu11==10.9.0.58
+six==1.16.0
+pydantic==2.9.2
+click==8.1.7
+nest-asyncio==1.5.6
+gmpy2==2.1.0
+matplotlib==3.8.2
+scipy==1.11.4
+typing_extensions==4.12.2
+statsmodels==0.14.0
+huggingface-hub==0.25.0
+frozenlist==1.4.1
+gpustat==1.1.1
+nvidia-nvtx-cu11==11.7.91
+safetensors==0.4.5
+stanza==1.9.2
+decorator==5.1.1
+seaborn==0.13.0
+sentencepiece==0.2.0
+PyYAML==6.0.1
+black==24.8.0
+protobuf==4.25.1
+pickleshare==0.7.5
+peft==0.13.0
+triton==2.0.0
+nvidia-cuda-runtime-cu11==11.7.99
+Jinja2==3.1.2
+nvidia-cusolver-cu11==11.4.0.1
+executing==1.2.0
+jupyter_client==8.1.0
+pluggy==1.3.0
+cmake==3.30.3
+pytz==2023.3.post1
+aiohappyeyeballs==2.4.2
+kiwisolver==1.4.5
+py-cpuinfo==9.0.0
+Pillow==10.1.0
+ptyprocess==0.7.0
+importlib_resources==6.4.5
+GitPython==3.1.43
+importlib-metadata==6.0.0
+iniconfig==2.0.0
+scikit-learn==1.3.2
+exceptiongroup==1.1.0
+networkx==2.8.6
+accelerate==1.0.0
+nltk==3.8.1
+shutilwhich==1.1.0
+fonttools==4.45.1
+future==0.18.3
+aiohttp==3.10.6
+wcwidth==0.2.5
+idna==3.6
+filelock==3.12.2
+pathspec==0.12.1
+jupyter_core==5.1.0
+lit==18.1.8
+nvidia-curand-cu11==10.2.10.91
+nvidia-cublas-cu11==11.10.3.66
+nvidia-ml-py==12.560.30
+msgpack==1.1.0
+python-dateutil==2.8.2
+blessed==1.20.0
+packaging==23.0
+gitdb==4.0.11
+yarl==1.13.0
+emoji==2.8.0
+tzdata==2023.3
+cycler==0.12.1
+tornado==6.2
+backcall==0.2.0
+plotnine==0.12.4
+ninja==1.11.1.1
+latex==0.7.0
+wandb==0.18.5
+setproctitle==1.3.3
+threadpoolctl==3.2.0
+requests==2.32.3
+pyparsing==3.1.1
+smmap==5.0.1
+pyzmq==23.0.0
+async-timeout==4.0.3
+annotated-types==0.7.0
+matplotlib-inline==0.1.6
+latexcodec==1.0.0
+ipython==8.0.0
+patsy==0.5.3
+contourpy==1.2.0
+multidict==6.1.0
+mizani==0.9.3
+urllib3==2.1.0
+tokenizers==0.20.0
+MarkupSafe==2.1.2
+pip==24.2
+pexpect==4.8.0
+tqdm==4.66.5
+jedi==0.18.2
+pydantic_core==2.23.4
+tempdir==0.7.1
+mpmath==1.2.1
+setuptools==72.1.0
+pytest==7.4.3
+pure-eval==0.2.2
+psutil==5.9.1
+comm==0.1.2
+nvidia-cuda-cupti-cu11==11.7.101
+nvidia-cuda-nvrtc-cu11==11.7.99
+regex==2023.10.3
+platformdirs==2.5.2
+asttokens==2.2.1
+torch==2.0.0
+nvidia-nccl-cu11==2.14.3
+xxhash==3.5.0

wandb/run-20241101_200502-8iosiep6/files/wandb-metadata.json

@@ -0,0 +1,97 @@
+{
+  "os":  "Linux-5.4.0-162-generic-x86_64-with-glibc2.31",
+  "python":  "3.9.19",
+  "startedAt":  "2024-11-02T00:05:02.629495Z",
+  "args":  [
+    "--perturbation",
+    "shuffle_nondeterministic",
+    "--train_set",
+    "10M",
+    "--batch_size",
+    "3",
+    "--epoch",
+    "3",
+    "--seed",
+    "0"
+  ],
+  "program":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py",
+  "codePath":  "train/train_deep_wandb.py",
+  "git":  {
+    "remote":  "git@hf.co:Yaning1001/Impossible_llm.git",
+    "commit":  "ed716cdcfcdea02b67f7ed0f3504c2b1c8b737c4"
+  },
+  "email":  "yaning1001@gmail.com",
+  "root":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train",
+  "host":  "mms-large-2",
+  "username":  "chunhui",
+  "executable":  "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/bin/python",
+  "codePathLocal":  "train_deep_wandb.py",
+  "cpu_count":  32,
+  "cpu_count_logical":  64,
+  "gpu":  "NVIDIA RTX A6000",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1888559353856",
+      "used":  "1754801463296"
+    }
+  },
+  "memory":  {
+    "total":  "202617098240"
+  },
+  "cpu":  {
+    "count":  32,
+    "countLogical":  64
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    }
+  ],
+  "cudaVersion":  "11.8"
+}

wandb/run-20241101_200502-8iosiep6/logs/debug-internal.log

@@ -0,0 +1,8 @@
+{"time":"2024-11-01T20:05:02.634314791-04:00","level":"INFO","msg":"using version","core version":"0.18.5"}
+{"time":"2024-11-01T20:05:02.634334811-04:00","level":"INFO","msg":"created symlink","path":"/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241101_200502-8iosiep6/logs/debug-core.log"}
+{"time":"2024-11-01T20:05:02.748799429-04:00","level":"INFO","msg":"created new stream","id":"8iosiep6"}
+{"time":"2024-11-01T20:05:02.74887203-04:00","level":"INFO","msg":"stream: started","id":"8iosiep6"}
+{"time":"2024-11-01T20:05:02.74889853-04:00","level":"INFO","msg":"handler: started","stream_id":{"value":"8iosiep6"}}
+{"time":"2024-11-01T20:05:02.749308363-04:00","level":"INFO","msg":"sender: started","stream_id":"8iosiep6"}
+{"time":"2024-11-01T20:05:02.749308913-04:00","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"8iosiep6"}}
+{"time":"2024-11-01T20:05:02.991863667-04:00","level":"INFO","msg":"Starting system monitor"}

wandb/run-20241101_200502-8iosiep6/logs/debug.log

@@ -0,0 +1,26 @@
+2024-11-01 20:05:02,625 INFO    MainThread:869509 [wandb_setup.py:_flush():79] Current SDK version is 0.18.5
+2024-11-01 20:05:02,625 INFO    MainThread:869509 [wandb_setup.py:_flush():79] Configure stats pid to 869509
+2024-11-01 20:05:02,625 INFO    MainThread:869509 [wandb_setup.py:_flush():79] Loading settings from /home/chunhui/.config/wandb/settings
+2024-11-01 20:05:02,625 INFO    MainThread:869509 [wandb_setup.py:_flush():79] Loading settings from /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/settings
+2024-11-01 20:05:02,625 INFO    MainThread:869509 [wandb_setup.py:_flush():79] Loading settings from environment variables: {}
+2024-11-01 20:05:02,625 INFO    MainThread:869509 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': None, '_disable_service': None}
+2024-11-01 20:05:02,625 INFO    MainThread:869509 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'train/train_deep_wandb.py', 'program_abspath': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py', 'program': '/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py'}
+2024-11-01 20:05:02,625 INFO    MainThread:869509 [wandb_setup.py:_flush():79] Applying login settings: {}
+2024-11-01 20:05:02,626 INFO    MainThread:869509 [wandb_init.py:_log_setup():534] Logging user logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241101_200502-8iosiep6/logs/debug.log
+2024-11-01 20:05:02,626 INFO    MainThread:869509 [wandb_init.py:_log_setup():535] Logging internal logs to /mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241101_200502-8iosiep6/logs/debug-internal.log
+2024-11-01 20:05:02,626 INFO    MainThread:869509 [wandb_init.py:init():621] calling init triggers
+2024-11-01 20:05:02,626 INFO    MainThread:869509 [wandb_init.py:init():628] wandb.init called with sweep_config: {}
+config: {}
+2024-11-01 20:05:02,626 INFO    MainThread:869509 [wandb_init.py:init():671] starting backend
+2024-11-01 20:05:02,626 INFO    MainThread:869509 [wandb_init.py:init():675] sending inform_init request
+2024-11-01 20:05:02,628 INFO    MainThread:869509 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2024-11-01 20:05:02,629 INFO    MainThread:869509 [wandb_init.py:init():688] backend started and connected
+2024-11-01 20:05:02,631 INFO    MainThread:869509 [wandb_init.py:init():783] updated telemetry
+2024-11-01 20:05:02,654 INFO    MainThread:869509 [wandb_init.py:init():816] communicating run to backend with 90.0 second timeout
+2024-11-01 20:05:02,984 INFO    MainThread:869509 [wandb_init.py:init():867] starting run threads in backend
+2024-11-01 20:05:03,124 INFO    MainThread:869509 [wandb_run.py:_console_start():2463] atexit reg
+2024-11-01 20:05:03,124 INFO    MainThread:869509 [wandb_run.py:_redirect():2311] redirect: wrap_raw
+2024-11-01 20:05:03,124 INFO    MainThread:869509 [wandb_run.py:_redirect():2376] Wrapping output streams.
+2024-11-01 20:05:03,124 INFO    MainThread:869509 [wandb_run.py:_redirect():2401] Redirects installed.
+2024-11-01 20:05:03,125 INFO    MainThread:869509 [wandb_init.py:init():911] run started, returning control to user process
+2024-11-01 20:05:03,126 INFO    MainThread:869509 [wandb_run.py:_config_callback():1390] config_cb None None {'perturbation': 'shuffle_nondeterministic', 'train_set': '10M', 'batch_size': 3, 'epoch': 3, 'seed': 0, 'lr': 5e-06}

wandb/run-20241101_201708-b4wkk29o/files/requirements.txt

@@ -0,0 +1,147 @@
+funcsigs==1.0.2
+sentry-sdk==2.17.0
+multiprocess==0.70.16
+numpy==1.26.2
+pluralizer==1.2.0
+debugpy==1.6.7
+nvidia-cudnn-cu11==8.5.0.96
+deepspeed==0.15.2
+data==0.4
+pandas==2.1.3
+tomli==2.0.1
+charset-normalizer==3.3.2
+attrs==24.2.0
+aiosignal==1.3.1
+fsspec==2023.10.0
+nvidia-cusparse-cu11==11.7.4.91
+zipp==3.12.0
+mypy-extensions==1.0.0
+datasets==3.0.1
+joblib==1.3.2
+hjson==3.1.0
+traitlets==5.7.1
+stack-data==0.6.0
+transformers==4.45.1
+sympy==1.11.1
+Pygments==2.15.0
+docker-pycreds==0.4.0
+dill==0.3.8
+wheel==0.44.0
+prompt-toolkit==3.0.30
+parso==0.8.3
+ipykernel==6.23.1
+pyarrow==17.0.0
+certifi==2023.11.17
+nvidia-cufft-cu11==10.9.0.58
+six==1.16.0
+pydantic==2.9.2
+click==8.1.7
+nest-asyncio==1.5.6
+gmpy2==2.1.0
+matplotlib==3.8.2
+scipy==1.11.4
+typing_extensions==4.12.2
+statsmodels==0.14.0
+huggingface-hub==0.25.0
+frozenlist==1.4.1
+gpustat==1.1.1
+nvidia-nvtx-cu11==11.7.91
+safetensors==0.4.5
+stanza==1.9.2
+decorator==5.1.1
+seaborn==0.13.0
+sentencepiece==0.2.0
+PyYAML==6.0.1
+black==24.8.0
+protobuf==4.25.1
+pickleshare==0.7.5
+peft==0.13.0
+triton==2.0.0
+nvidia-cuda-runtime-cu11==11.7.99
+Jinja2==3.1.2
+nvidia-cusolver-cu11==11.4.0.1
+executing==1.2.0
+jupyter_client==8.1.0
+pluggy==1.3.0
+cmake==3.30.3
+pytz==2023.3.post1
+aiohappyeyeballs==2.4.2
+kiwisolver==1.4.5
+py-cpuinfo==9.0.0
+Pillow==10.1.0
+ptyprocess==0.7.0
+importlib_resources==6.4.5
+GitPython==3.1.43
+importlib-metadata==6.0.0
+iniconfig==2.0.0
+scikit-learn==1.3.2
+exceptiongroup==1.1.0
+networkx==2.8.6
+accelerate==1.0.0
+nltk==3.8.1
+shutilwhich==1.1.0
+fonttools==4.45.1
+future==0.18.3
+aiohttp==3.10.6
+wcwidth==0.2.5
+idna==3.6
+filelock==3.12.2
+pathspec==0.12.1
+jupyter_core==5.1.0
+lit==18.1.8
+nvidia-curand-cu11==10.2.10.91
+nvidia-cublas-cu11==11.10.3.66
+nvidia-ml-py==12.560.30
+msgpack==1.1.0
+python-dateutil==2.8.2
+blessed==1.20.0
+packaging==23.0
+gitdb==4.0.11
+yarl==1.13.0
+emoji==2.8.0
+tzdata==2023.3
+cycler==0.12.1
+tornado==6.2
+backcall==0.2.0
+plotnine==0.12.4
+ninja==1.11.1.1
+latex==0.7.0
+wandb==0.18.5
+setproctitle==1.3.3
+threadpoolctl==3.2.0
+requests==2.32.3
+pyparsing==3.1.1
+smmap==5.0.1
+pyzmq==23.0.0
+async-timeout==4.0.3
+annotated-types==0.7.0
+matplotlib-inline==0.1.6
+latexcodec==1.0.0
+ipython==8.0.0
+patsy==0.5.3
+contourpy==1.2.0
+multidict==6.1.0
+mizani==0.9.3
+urllib3==2.1.0
+tokenizers==0.20.0
+MarkupSafe==2.1.2
+pip==24.2
+pexpect==4.8.0
+tqdm==4.66.5
+jedi==0.18.2
+pydantic_core==2.23.4
+tempdir==0.7.1
+mpmath==1.2.1
+setuptools==72.1.0
+pytest==7.4.3
+pure-eval==0.2.2
+psutil==5.9.1
+comm==0.1.2
+nvidia-cuda-cupti-cu11==11.7.101
+nvidia-cuda-nvrtc-cu11==11.7.99
+regex==2023.10.3
+platformdirs==2.5.2
+asttokens==2.2.1
+torch==2.0.0
+nvidia-nccl-cu11==2.14.3
+xxhash==3.5.0

wandb/run-20241101_201708-b4wkk29o/logs/debug-internal.log

@@ -0,0 +1,8 @@
+{"time":"2024-11-01T20:17:08.116255333-04:00","level":"INFO","msg":"using version","core version":"0.18.5"}
+{"time":"2024-11-01T20:17:08.116269143-04:00","level":"INFO","msg":"created symlink","path":"/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/wandb/run-20241101_201708-b4wkk29o/logs/debug-core.log"}
+{"time":"2024-11-01T20:17:08.222816277-04:00","level":"INFO","msg":"created new stream","id":"b4wkk29o"}
+{"time":"2024-11-01T20:17:08.222884707-04:00","level":"INFO","msg":"stream: started","id":"b4wkk29o"}
+{"time":"2024-11-01T20:17:08.222956257-04:00","level":"INFO","msg":"sender: started","stream_id":"b4wkk29o"}
+{"time":"2024-11-01T20:17:08.222931577-04:00","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"b4wkk29o"}}
+{"time":"2024-11-01T20:17:08.223042088-04:00","level":"INFO","msg":"handler: started","stream_id":{"value":"b4wkk29o"}}
+{"time":"2024-11-01T20:17:09.978983673-04:00","level":"INFO","msg":"Starting system monitor"}

wandb/run-20241101_202058-ptl7coag/files/wandb-metadata.json

@@ -0,0 +1,97 @@
+{
+  "os":  "Linux-5.4.0-162-generic-x86_64-with-glibc2.31",
+  "python":  "3.9.19",
+  "startedAt":  "2024-11-02T00:20:58.449642Z",
+  "args":  [
+    "--perturbation",
+    "shuffle_nondeterministic",
+    "--train_set",
+    "10M",
+    "--batch_size",
+    "3",
+    "--epoch",
+    "3",
+    "--seed",
+    "0"
+  ],
+  "program":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train/train_deep_wandb.py",
+  "codePath":  "train/train_deep_wandb.py",
+  "git":  {
+    "remote":  "git@hf.co:Yaning1001/Impossible_llm.git",
+    "commit":  "ed716cdcfcdea02b67f7ed0f3504c2b1c8b737c4"
+  },
+  "email":  "yaning1001@gmail.com",
+  "root":  "/mnt/ssd3/chunhui/yaning/project/impossible_llm/train",
+  "host":  "mms-large-2",
+  "username":  "chunhui",
+  "executable":  "/mnt/ssd3/chunhui/miniconda/envs/impossible_llm/bin/python",
+  "codePathLocal":  "train_deep_wandb.py",
+  "cpu_count":  32,
+  "cpu_count_logical":  64,
+  "gpu":  "NVIDIA RTX A6000",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1888559353856",
+      "used":  "1754804830208"
+    }
+  },
+  "memory":  {
+    "total":  "202617098240"
+  },
+  "cpu":  {
+    "count":  32,
+    "countLogical":  64
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    },
+    {
+      "name":  "NVIDIA RTX A6000",
+      "memoryTotal":  "51527024640",
+      "cudaCores":  10752,
+      "architecture":  "Ampere"
+    }
+  ],
+  "cudaVersion":  "11.8"
+}