submission_id: nvidia-llama3-chatqa-1-5-70b_v1
developer_uid: windyheath
alignment_samples: 0
best_of: 4
celo_rating: 1153.17
display_name: nvidia-llama3-chatqa-1-5-70b_v1
formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 4, 'max_output_tokens': 64}
ineligible_reason: propriety_total_count < 800
is_internal_developer: False
language_model: nvidia/Llama3-ChatQA-1.5-70B
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_eval_status: success
model_group: nvidia/Llama3-ChatQA-1.5
model_name: nvidia-llama3-chatqa-1-5-70b_v1
model_num_parameters: 70553706496.0
model_repo: nvidia/Llama3-ChatQA-1.5-70B
model_size: 71B
num_battles: 10452
num_wins: 4792
propriety_score: 0.7019108280254777
propriety_total_count: 785.0
ranking_group: single
reward_formatter: {'bot_template': '{bot_name}: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '{user_name}: {message}\n'}
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
status: torndown
submission_type: basic
timestamp: 2024-06-07T21:27:32+00:00
us_pacific_date: 2024-06-07
win_ratio: 0.458476846536548
Resubmit model
Running pipeline stage MKMLizer
Starting job with name nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer
Waiting for job on nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer to finish
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║     _____            __           __                                ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║       /___/                                                         ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║                                                                     ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║  Version: 0.8.14                                                    ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║  https://mk1.ai                                                     ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║                                                                     ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║  belonging to:                                                      ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║                                                                     ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║  Chai Research Corp.                                                ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ║                                                                     ║
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/utils/_deprecation.py:131: FutureWarning: 'list_files_info' (from 'huggingface_hub.hf_api') is deprecated and will be removed from version '0.23'. Use `list_repo_tree` and `get_paths_info` instead.
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer:   warnings.warn(warning_message, FutureWarning)
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: Downloaded to shared memory in 231.407s
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: quantizing model to /dev/shm/model_cache
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: 
Loading 0:   0%|          | 0/803 [00:00<?, ?it/s]
Loading 0:   0%|          | 4/803 [00:00<00:26, 29.82it/s]
Loading 0:   1%|          | 9/803 [00:00<00:20, 38.63it/s]
Loading 0:   2%|▏         | 14/803 [00:00<00:19, 40.35it/s]
Loading 0:   2%|▏         | 19/803 [00:00<00:17, 43.75it/s]
Loading 0:   3%|▎         | 24/803 [00:00<00:18, 43.04it/s]
Loading 0:   4%|▍         | 32/803 [00:00<00:14, 54.51it/s]
Loading 0:   5%|▍         | 38/803 [00:00<00:17, 43.88it/s]
Loading 0:   6%|▌         | 47/803 [00:00<00:14, 53.25it/s]
Loading 0:   7%|▋         | 53/803 [00:01<00:31, 23.84it/s]
Loading 0:   7%|▋         | 58/803 [00:01<00:29, 25.55it/s]
Loading 0:   8%|▊         | 62/803 [00:01<00:32, 22.86it/s]
Loading 0:   9%|▊         | 70/803 [00:02<00:23, 31.08it/s]
Loading 0:   9%|▉         | 75/803 [00:02<00:22, 32.11it/s]
Loading 0:  10%|█         | 81/803 [00:02<00:20, 34.65it/s]
Loading 0:  11%|█         | 87/803 [00:02<00:18, 39.71it/s]
Loading 0:  11%|█▏        | 92/803 [00:02<00:19, 36.14it/s]
Loading 0:  13%|█▎        | 101/803 [00:02<00:16, 42.85it/s]
Loading 0:  13%|█▎        | 106/803 [00:02<00:15, 44.01it/s]
Loading 0:  14%|█▍        | 111/803 [00:03<00:37, 18.70it/s]
Loading 0:  14%|█▍        | 115/803 [00:03<00:35, 19.22it/s]
Loading 0:  15%|█▌        | 123/803 [00:03<00:24, 27.41it/s]
Loading 0:  16%|█▌        | 128/803 [00:04<00:23, 28.86it/s]
Loading 0:  17%|█▋        | 134/803 [00:04<00:20, 31.96it/s]
Loading 0:  17%|█▋        | 139/803 [00:15<07:01,  1.58it/s]
Loading 0:  18%|█▊        | 144/803 [00:15<05:07,  2.14it/s]
Loading 0:  19%|█▊        | 149/803 [00:15<03:42,  2.94it/s]
Loading 0:  19%|█▉        | 154/803 [00:15<02:42,  4.00it/s]
Loading 0:  20%|█▉        | 159/803 [00:16<01:58,  5.44it/s]
Loading 0:  20%|██        | 164/803 [00:16<01:50,  5.76it/s]
Loading 0:  21%|██        | 168/803 [00:16<01:28,  7.19it/s]
Loading 0:  22%|██▏       | 173/803 [00:17<01:04,  9.74it/s]
Loading 0:  22%|██▏       | 177/803 [00:17<00:53, 11.69it/s]
Loading 0:  23%|██▎       | 183/803 [00:17<00:37, 16.32it/s]
Loading 0:  23%|██▎       | 187/803 [00:17<00:33, 18.14it/s]
Loading 0:  24%|██▍       | 193/803 [00:17<00:25, 23.87it/s]
Loading 0:  25%|██▍       | 198/803 [00:17<00:23, 25.49it/s]
Loading 0:  25%|██▌       | 204/803 [00:17<00:20, 28.93it/s]
Loading 0:  26%|██▌       | 209/803 [00:18<00:18, 32.12it/s]
Loading 0:  27%|██▋       | 214/803 [00:18<00:17, 33.21it/s]
Loading 0:  28%|██▊       | 221/803 [00:18<00:32, 18.01it/s]
Loading 0:  28%|██▊       | 225/803 [00:19<00:31, 18.23it/s]
Loading 0:  29%|██▉       | 233/803 [00:19<00:22, 25.80it/s]
Loading 0:  30%|██▉       | 237/803 [00:19<00:21, 26.23it/s]
Loading 0:  30%|███       | 243/803 [00:19<00:17, 31.79it/s]
Loading 0:  31%|███       | 248/803 [00:19<00:17, 31.31it/s]
Loading 0:  32%|███▏      | 254/803 [00:19<00:16, 32.91it/s]
Loading 0:  32%|███▏      | 258/803 [00:19<00:16, 33.76it/s]
Loading 0:  33%|███▎      | 263/803 [00:19<00:14, 36.62it/s]
Loading 0:  33%|███▎      | 268/803 [00:20<00:17, 30.96it/s]
Loading 0:  34%|███▍      | 273/803 [00:20<00:15, 34.25it/s]
Loading 0:  35%|███▍      | 280/803 [00:21<00:32, 16.15it/s]
Loading 0:  35%|███▌      | 283/803 [00:21<00:31, 16.45it/s]
Loading 0:  36%|███▌      | 286/803 [00:21<00:30, 16.86it/s]
Loading 0:  36%|███▋      | 293/803 [00:21<00:21, 23.59it/s]
Loading 0:  37%|███▋      | 297/803 [00:33<06:31,  1.29it/s]
Loading 0:  38%|███▊      | 303/803 [00:33<04:14,  1.96it/s]
Loading 0:  38%|███▊      | 307/803 [00:33<03:14,  2.55it/s]
Loading 0:  39%|███▉      | 313/803 [00:33<02:08,  3.82it/s]
Loading 0:  40%|███▉      | 318/803 [00:33<01:34,  5.14it/s]
Loading 0:  40%|████      | 324/803 [00:34<01:06,  7.22it/s]
Loading 0:  41%|████      | 329/803 [00:34<00:50,  9.46it/s]
Loading 0:  42%|████▏     | 339/803 [00:35<00:43, 10.60it/s]
Loading 0:  43%|████▎     | 343/803 [00:35<00:39, 11.63it/s]
Loading 0:  43%|████▎     | 346/803 [00:35<00:36, 12.63it/s]
Loading 0:  44%|████▍     | 353/803 [00:35<00:25, 17.96it/s]
Loading 0:  44%|████▍     | 357/803 [00:35<00:23, 18.86it/s]
Loading 0:  45%|████▌     | 363/803 [00:35<00:18, 24.22it/s]
Loading 0:  46%|████▌     | 367/803 [00:35<00:17, 24.79it/s]
Loading 0:  46%|████▋     | 373/803 [00:36<00:13, 30.72it/s]
Loading 0:  47%|████▋     | 378/803 [00:36<00:13, 30.63it/s]
Loading 0:  48%|████▊     | 384/803 [00:36<00:12, 32.67it/s]
Loading 0:  48%|████▊     | 389/803 [00:36<00:11, 35.17it/s]
Loading 0:  49%|████▉     | 394/803 [00:37<00:23, 17.23it/s]
Loading 0:  50%|████▉     | 398/803 [00:37<00:21, 19.01it/s]
Loading 0:  50%|█████     | 403/803 [00:37<00:17, 23.30it/s]
Loading 0:  51%|█████     | 407/803 [00:37<00:16, 23.66it/s]
Loading 0:  51%|█████▏    | 413/803 [00:37<00:13, 29.83it/s]
Loading 0:  52%|█████▏    | 417/803 [00:37<00:13, 29.22it/s]
Loading 0:  53%|█████▎    | 423/803 [00:37<00:10, 35.44it/s]
Loading 0:  53%|█████▎    | 428/803 [00:38<00:10, 34.30it/s]
Loading 0:  54%|█████▍    | 434/803 [00:38<00:10, 36.08it/s]
Loading 0:  55%|█████▍    | 439/803 [00:38<00:09, 38.15it/s]
Loading 0:  55%|█████▌    | 444/803 [00:38<00:09, 37.72it/s]
Loading 0:  56%|█████▌    | 449/803 [00:50<00:09, 37.72it/s]
Loading 0:  56%|█████▌    | 450/803 [00:50<03:57,  1.49it/s]
Loading 0:  56%|█████▌    | 451/803 [00:50<03:56,  1.49it/s]
Loading 0:  57%|█████▋    | 454/803 [00:51<03:04,  1.90it/s]
Loading 0:  57%|█████▋    | 457/803 [00:51<02:20,  2.47it/s]
Loading 0:  58%|█████▊    | 463/803 [00:51<01:23,  4.09it/s]
Loading 0:  58%|█████▊    | 466/803 [00:51<01:07,  4.96it/s]
Loading 0:  59%|█████▉    | 473/803 [00:51<00:40,  8.14it/s]
Loading 0:  59%|█████▉    | 477/803 [00:51<00:33,  9.83it/s]
Loading 0:  60%|██████    | 483/803 [00:52<00:23, 13.76it/s]
Loading 0:  61%|██████    | 487/803 [00:52<00:20, 15.41it/s]
Loading 0:  61%|██████▏   | 493/803 [00:52<00:15, 20.30it/s]
Loading 0:  62%|██████▏   | 497/803 [00:52<00:14, 20.76it/s]
Loading 0:  63%|██████▎   | 503/803 [00:52<00:11, 25.53it/s]
Loading 0:  64%|██████▎   | 510/803 [00:53<00:20, 14.55it/s]
Loading 0:  64%|██████▍   | 513/803 [00:53<00:19, 14.89it/s]
Loading 0:  64%|██████▍   | 516/803 [00:53<00:19, 15.08it/s]
Loading 0:  65%|██████▌   | 523/803 [00:54<00:13, 20.94it/s]
Loading 0:  66%|██████▌   | 526/803 [00:54<00:14, 19.62it/s]
Loading 0:  66%|██████▋   | 533/803 [00:54<00:10, 26.32it/s]
Loading 0:  67%|██████▋   | 537/803 [00:54<00:10, 24.78it/s]
Loading 0:  68%|██████▊   | 543/803 [00:54<00:08, 29.87it/s]
Loading 0:  68%|██████▊   | 547/803 [00:54<00:09, 27.60it/s]
Loading 0:  69%|██████▉   | 553/803 [00:54<00:07, 32.34it/s]
Loading 0:  69%|██████▉   | 557/803 [00:55<00:08, 28.44it/s]
Loading 0:  70%|███████   | 566/803 [00:55<00:05, 40.15it/s]
Loading 0:  71%|███████   | 571/803 [00:56<00:14, 15.94it/s]
Loading 0:  72%|███████▏  | 575/803 [00:56<00:14, 15.91it/s]
Loading 0:  73%|███████▎  | 583/803 [00:56<00:09, 22.05it/s]
Loading 0:  73%|███████▎  | 587/803 [00:56<00:09, 22.08it/s]
Loading 0:  74%|███████▍  | 593/803 [00:56<00:07, 26.26it/s]
Loading 0:  74%|███████▍  | 597/803 [00:56<00:08, 25.46it/s]
Loading 0:  75%|███████▌  | 603/803 [01:08<02:18,  1.45it/s]
Loading 0:  75%|███████▌  | 604/803 [01:09<02:08,  1.55it/s]
Loading 0:  76%|███████▌  | 607/803 [01:09<01:38,  2.00it/s]
Loading 0:  76%|███████▋  | 613/803 [01:09<00:57,  3.28it/s]
Loading 0:  77%|███████▋  | 617/803 [01:09<00:43,  4.31it/s]
Loading 0:  78%|███████▊  | 624/803 [01:10<00:32,  5.55it/s]
Loading 0:  78%|███████▊  | 627/803 [01:10<00:27,  6.40it/s]
Loading 0:  79%|███████▉  | 633/803 [01:10<00:18,  9.25it/s]
Loading 0:  79%|███████▉  | 636/803 [01:10<00:16, 10.17it/s]
Loading 0:  80%|████████  | 643/803 [01:10<00:10, 14.95it/s]
Loading 0:  81%|████████  | 647/803 [01:11<00:09, 16.05it/s]
Loading 0:  81%|████████▏ | 653/803 [01:11<00:07, 20.22it/s]
Loading 0:  82%|████████▏ | 657/803 [01:11<00:07, 20.60it/s]
Loading 0:  83%|████████▎ | 663/803 [01:11<00:05, 24.79it/s]
Loading 0:  83%|████████▎ | 667/803 [01:11<00:05, 24.46it/s]
Loading 0:  84%|████████▍ | 673/803 [01:11<00:04, 28.90it/s]
Loading 0:  84%|████████▍ | 677/803 [01:11<00:04, 30.53it/s]
Loading 0:  85%|████████▍ | 681/803 [01:12<00:08, 14.51it/s]
Loading 0:  85%|████████▌ | 684/803 [01:12<00:08, 14.85it/s]
Loading 0:  86%|████████▌ | 688/803 [01:12<00:06, 17.62it/s]
Loading 0:  86%|████████▋ | 693/803 [01:13<00:05, 21.74it/s]
Loading 0:  87%|████████▋ | 696/803 [01:13<00:05, 20.50it/s]
Loading 0:  88%|████████▊ | 703/803 [01:13<00:03, 27.56it/s]
Loading 0:  88%|████████▊ | 707/803 [01:13<00:03, 25.72it/s]
Loading 0:  89%|████████▉ | 713/803 [01:13<00:02, 30.38it/s]
Loading 0:  89%|████████▉ | 717/803 [01:13<00:03, 27.88it/s]
Loading 0:  90%|█████████ | 723/803 [01:13<00:02, 31.17it/s]
Loading 0:  91%|█████████ | 727/803 [01:14<00:02, 28.84it/s]
Loading 0:  91%|█████████▏| 733/803 [01:14<00:02, 33.72it/s]
Loading 0:  92%|█████████▏| 740/803 [01:14<00:03, 17.91it/s]
Loading 0:  93%|█████████▎| 743/803 [01:15<00:03, 17.50it/s]
Loading 0:  93%|█████████▎| 746/803 [01:15<00:03, 17.26it/s]
Loading 0:  94%|█████████▍| 753/803 [01:15<00:02, 23.79it/s]
Loading 0:  94%|█████████▍| 757/803 [01:27<00:36,  1.25it/s]
Loading 0:  95%|█████████▌| 763/803 [01:27<00:21,  1.89it/s]
Loading 0:  96%|█████████▌| 767/803 [01:28<00:14,  2.45it/s]
Loading 0:  96%|█████████▋| 773/803 [01:28<00:08,  3.63it/s]
Loading 0:  97%|█████████▋| 777/803 [01:28<00:05,  4.59it/s]
Loading 0:  98%|█████████▊| 783/803 [01:28<00:02,  6.71it/s]
Loading 0:  98%|█████████▊| 787/803 [01:28<00:01,  8.08it/s]
Loading 0:  99%|█████████▉| 796/803 [01:28<00:00, 13.55it/s]
Loading 0:  99%|█████████▉| 798/803 [01:39<00:00, 13.55it/s]
Loading 0: 100%|█████████▉| 799/803 [01:39<00:02,  1.45it/s]
Loading 0: 100%|█████████▉| 801/803 [01:39<00:01,  1.65it/s]
                                                            
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: quantized model in 114.075s
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: Processed model nvidia/Llama3-ChatQA-1.5-70B in 356.790s
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: creating bucket guanaco-mkml-models
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/nvidia-llama3-chatqa-1-5-70b-v1
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/nvidia-llama3-chatqa-1-5-70b-v1/tokenizer_config.json
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/nvidia-llama3-chatqa-1-5-70b-v1/config.json
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/nvidia-llama3-chatqa-1-5-70b-v1/tokenizer.json
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/nvidia-llama3-chatqa-1-5-70b-v1/special_tokens_map.json
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.5.safetensors s3://guanaco-mkml-models/nvidia-llama3-chatqa-1-5-70b-v1/flywheel_model.5.safetensors
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/nvidia-llama3-chatqa-1-5-70b-v1/flywheel_model.0.safetensors
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:913: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer:   warnings.warn(
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:757: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer:   warnings.warn(
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:468: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer:   warnings.warn(
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly.  To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer:   return self.fget.__get__(instance, owner)()
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: Saving duration: 0.260s
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.990s
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: creating bucket guanaco-reward-models
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/nvidia-llama3-chatqa-1-5-70b-v1_reward
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/nvidia-llama3-chatqa-1-5-70b-v1_reward/merges.txt
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/nvidia-llama3-chatqa-1-5-70b-v1_reward/special_tokens_map.json
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/nvidia-llama3-chatqa-1-5-70b-v1_reward/config.json
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/nvidia-llama3-chatqa-1-5-70b-v1_reward/vocab.json
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/nvidia-llama3-chatqa-1-5-70b-v1_reward/tokenizer_config.json
nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/nvidia-llama3-chatqa-1-5-70b-v1_reward/tokenizer.json
Job nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer completed after 405.56s with status: succeeded
Stopping job with name nvidia-llama3-chatqa-1-5-70b-v1-mkmlizer
Pipeline stage MKMLizer completed in 406.01s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service nvidia-llama3-chatqa-1-5-70b-v1
Waiting for inference service nvidia-llama3-chatqa-1-5-70b-v1 to be ready
Inference service nvidia-llama3-chatqa-1-5-70b-v1 ready after 261.43055295944214s
Pipeline stage ISVCDeployer completed in 267.02s
Running pipeline stage StressChecker
Received healthy response to inference request in 4.0315117835998535s
Received healthy response to inference request in 4.222510099411011s
Received healthy response to inference request in 4.190370559692383s
Received healthy response to inference request in 4.233764886856079s
Received healthy response to inference request in 3.6583139896392822s
5 requests
0 failed requests
5th percentile: 3.7329535484313965
10th percentile: 3.8075931072235107
20th percentile: 3.9568722248077393
30th percentile: 4.063283538818359
40th percentile: 4.126827049255371
50th percentile: 4.190370559692383
60th percentile: 4.203226375579834
70th percentile: 4.216082191467285
80th percentile: 4.224761056900024
90th percentile: 4.229262971878052
95th percentile: 4.231513929367066
99th percentile: 4.233314695358277
mean time: 4.067294263839722
Pipeline stage StressChecker completed in 21.17s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.04s
Running pipeline stage DaemonicSafetyScorer
Running M-Eval for topic stay_in_character
Pipeline stage DaemonicSafetyScorer completed in 0.04s
M-Eval Dataset for topic stay_in_character is loaded
nvidia-llama3-chatqa-1-5-70b_v1 status is now deployed due to DeploymentManager action
nvidia-llama3-chatqa-1-5-70b_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of nvidia-llama3-chatqa-1-5-70b_v1
Running pipeline stage ISVCDeleter
Checking if service nvidia-llama3-chatqa-1-5-70b-v1 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 3.93s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key nvidia-llama3-chatqa-1-5-70b-v1/config.json from bucket guanaco-mkml-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1/flywheel_model.1.safetensors from bucket guanaco-mkml-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1/flywheel_model.2.safetensors from bucket guanaco-mkml-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1/flywheel_model.3.safetensors from bucket guanaco-mkml-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1/flywheel_model.4.safetensors from bucket guanaco-mkml-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1/flywheel_model.5.safetensors from bucket guanaco-mkml-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key nvidia-llama3-chatqa-1-5-70b-v1_reward/config.json from bucket guanaco-reward-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key nvidia-llama3-chatqa-1-5-70b-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 12.59s
nvidia-llama3-chatqa-1-5-70b_v1 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics