rirv938-llama-8b-dpo-vs-_2870

submission_id: rirv938-llama-8b-dpo-vs-_2870_v1
developer_uid: end_to_end_test
best_of: 1
celo_rating: 1258.88
display_name: rirv938-llama-8b-dpo-vs-_2870_v1
family_friendly_score: 0.0
formatter: {'memory_template': '', 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '', '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': 256, 'best_of': 1, 'max_output_tokens': 1}
gpu_counts: {'NVIDIA RTX A5000': 1}
ineligible_reason: model is only for e2e test
is_internal_developer: True
language_model: rirv938/llama_8b_dpo_vs_dpo_250k_2535
latencies: [{'batch_size': 1, 'throughput': 0.10719560218310165, 'latency_mean': 9.328644860982894, 'latency_p50': 9.320436596870422, 'latency_p90': 9.444816160202027}]
max_input_tokens: 256
max_output_tokens: 1
model_architecture: LlamaForSequenceClassification
model_group: rirv938/llama_8b_dpo_vs_
model_name: rirv938-llama-8b-dpo-vs-_2870_v1
model_num_parameters: 8030261248.0
model_repo: rirv938/llama_8b_dpo_vs_dpo_250k_2535
model_size: 8B
num_battles: 10979
num_wins: 5578
ranking_group: single
status: torndown
submission_type: basic
timestamp: 2024-09-14T17:50:31+00:00
us_pacific_date: 2024-09-14
win_ratio: 0.5080608434283632
Download Preference Data
Resubmit model
Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Failed to get response for submission chaiml-llama-8b-pairwis_8189_v19: ('http://chaiml-llama-8b-pairwis-8189-v19-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'readfrom tcp 127.0.0.1:48956->127.0.0.1:8080: write tcp 127.0.0.1:48956->127.0.0.1:8080: use of closed network connection\n')
Starting job with name rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer
Waiting for job on rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer to finish
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║     _____            __           __                                ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║       /___/                                                         ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║                                                                     ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║                                                                     ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║  belonging to:                                                      ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║                                                                     ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ║                                                                     ║
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: Downloaded to shared memory in 32.638s
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:t0, folder:/tmp/tmpdp1wl3ai, device:0
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Failed to get response for submission blend_sanen_2024-09-09: ('http://zonemercy-lexical-nemo-1518-v18-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:35370->127.0.0.1:8080: read: connection reset by peer\n')
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: quantized model in 84.995s
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: Processed model rirv938/llama_8b_dpo_vs_dpo_250k_2535 in 117.633s
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: creating bucket guanaco-mkml-models
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rirv938-llama-8b-dpo-vs-2870-v1
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rirv938-llama-8b-dpo-vs-2870-v1/tokenizer.json
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rirv938-llama-8b-dpo-vs-2870-v1/flywheel_model.0.safetensors
rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   1%|          | 3/291 [00:00<00:56,  5.13it/s]
Loading 0:   1%|▏         | 4/291 [00:01<01:31,  3.14it/s]
Loading 0:   2%|▏         | 5/291 [00:01<02:02,  2.34it/s]
Loading 0:   3%|▎         | 8/291 [00:02<01:03,  4.47it/s]
Loading 0:   3%|▎         | 9/291 [00:02<01:02,  4.52it/s]
Loading 0:   3%|▎         | 10/291 [00:02<00:55,  5.02it/s]
Loading 0:   4%|▍         | 12/291 [00:02<01:05,  4.23it/s]
Loading 0:   4%|▍         | 13/291 [00:03<01:27,  3.19it/s]
Loading 0:   5%|▍         | 14/291 [00:04<01:49,  2.54it/s]
Loading 0:   6%|▌         | 17/291 [00:04<01:02,  4.36it/s]
Loading 0:   6%|▌         | 18/291 [00:04<00:59,  4.59it/s]
Loading 0:   7%|▋         | 19/291 [00:04<00:53,  5.09it/s]
Loading 0:   7%|▋         | 21/291 [00:05<01:02,  4.30it/s]
Loading 0:   8%|▊         | 22/291 [00:05<01:22,  3.24it/s]
Loading 0:   8%|▊         | 23/291 [00:06<01:44,  2.58it/s]
Loading 0:   9%|▉         | 26/291 [00:06<01:00,  4.39it/s]
Loading 0:   9%|▉         | 27/291 [00:06<00:57,  4.62it/s]
Loading 0:  10%|█         | 30/291 [00:07<00:57,  4.52it/s]
Loading 0:  11%|█         | 31/291 [00:08<01:14,  3.50it/s]
Loading 0:  11%|█         | 32/291 [00:08<01:32,  2.79it/s]
Loading 0:  12%|█▏        | 35/291 [00:09<00:57,  4.46it/s]
Loading 0:  12%|█▏        | 36/291 [00:09<00:54,  4.67it/s]
Loading 0:  13%|█▎        | 39/291 [00:09<00:55,  4.56it/s]
Loading 0:  14%|█▎        | 40/291 [00:10<01:10,  3.55it/s]
Loading 0:  14%|█▍        | 41/291 [00:11<01:28,  2.82it/s]
Loading 0:  15%|█▌        | 44/291 [00:11<00:54,  4.52it/s]
Loading 0:  15%|█▌        | 45/291 [00:11<00:52,  4.72it/s]
Loading 0:  16%|█▋        | 48/291 [00:12<00:52,  4.59it/s]
Loading 0:  17%|█▋        | 49/291 [00:12<01:08,  3.56it/s]
Loading 0:  17%|█▋        | 50/291 [00:13<01:25,  2.81it/s]
Loading 0:  18%|█▊        | 53/291 [00:13<00:53,  4.43it/s]
Loading 0:  19%|█▊        | 54/291 [00:13<00:50,  4.65it/s]
Loading 0:  19%|█▉        | 55/291 [00:13<00:45,  5.17it/s]
Loading 0:  20%|█▉        | 57/291 [00:14<00:53,  4.37it/s]
Loading 0:  20%|█▉        | 58/291 [00:15<01:10,  3.31it/s]
Loading 0:  20%|██        | 59/291 [00:15<01:28,  2.62it/s]
Loading 0:  21%|██▏       | 62/291 [00:15<00:52,  4.39it/s]
Loading 0:  22%|██▏       | 63/291 [00:16<00:49,  4.62it/s]
Loading 0:  22%|██▏       | 64/291 [00:16<00:43,  5.20it/s]
Loading 0:  23%|██▎       | 66/291 [00:16<00:51,  4.37it/s]
Loading 0:  23%|██▎       | 67/291 [00:17<01:07,  3.30it/s]
Loading 0:  23%|██▎       | 68/291 [00:17<01:25,  2.59it/s]
Loading 0:  24%|██▍       | 71/291 [00:18<00:49,  4.45it/s]
Loading 0:  25%|██▍       | 72/291 [00:18<00:46,  4.67it/s]
Loading 0:  26%|██▌       | 75/291 [00:19<00:47,  4.56it/s]
Loading 0:  26%|██▌       | 76/291 [00:19<01:01,  3.52it/s]
Loading 0:  26%|██▋       | 77/291 [00:20<01:16,  2.80it/s]
Loading 0:  27%|██▋       | 80/291 [00:20<00:47,  4.47it/s]
Loading 0:  28%|██▊       | 81/291 [00:20<00:44,  4.68it/s]
Loading 0:  28%|██▊       | 82/291 [00:20<00:40,  5.16it/s]
Loading 0:  29%|██▊       | 83/291 [00:20<00:39,  5.26it/s]
Loading 0:  29%|██▉       | 84/291 [00:21<00:59,  3.50it/s]
Loading 0:  29%|██▉       | 85/291 [00:22<01:14,  2.77it/s]
Loading 0:  30%|██▉       | 86/291 [00:22<01:29,  2.29it/s]
Loading 0:  31%|███       | 89/291 [00:22<00:47,  4.22it/s]
Loading 0:  31%|███       | 90/291 [00:23<00:44,  4.49it/s]
Loading 0:  31%|███▏      | 91/291 [00:23<00:39,  5.10it/s]
Loading 0:  32%|███▏      | 93/291 [00:23<00:46,  4.30it/s]
Loading 0:  32%|███▏      | 94/291 [00:24<01:00,  3.25it/s]
Loading 0:  33%|███▎      | 95/291 [00:24<01:16,  2.57it/s]
Loading 0:  34%|███▎      | 98/291 [00:25<00:43,  4.46it/s]
Loading 0:  34%|███▍      | 99/291 [00:25<00:40,  4.69it/s]
Loading 0:  35%|███▌      | 102/291 [00:26<00:41,  4.56it/s]
Loading 0:  35%|███▌      | 103/291 [00:26<00:53,  3.53it/s]
Loading 0:  36%|███▌      | 104/291 [00:27<01:06,  2.80it/s]
Loading 0:  37%|███▋      | 107/291 [00:27<00:40,  4.53it/s]
Loading 0:  37%|███▋      | 108/291 [00:27<00:38,  4.74it/s]
Loading 0:  37%|███▋      | 109/291 [00:27<00:34,  5.25it/s]
Loading 0:  38%|███▊      | 111/291 [00:28<00:41,  4.37it/s]
Loading 0:  38%|███▊      | 112/291 [00:28<00:54,  3.31it/s]
Loading 0:  39%|███▉      | 113/291 [00:29<01:08,  2.60it/s]
Loading 0:  40%|███▉      | 116/291 [00:29<00:39,  4.38it/s]
Loading 0:  40%|████      | 117/291 [00:29<00:37,  4.61it/s]
Loading 0:  41%|████      | 120/291 [00:30<00:37,  4.53it/s]
Loading 0:  42%|████▏     | 121/291 [00:31<00:48,  3.53it/s]
Loading 0:  42%|████▏     | 122/291 [00:31<01:00,  2.81it/s]
Loading 0:  43%|████▎     | 125/291 [00:32<00:37,  4.47it/s]
Loading 0:  43%|████▎     | 126/291 [00:32<00:35,  4.69it/s]
Loading 0:  44%|████▎     | 127/291 [00:32<00:31,  5.23it/s]
Loading 0:  44%|████▍     | 129/291 [00:32<00:36,  4.40it/s]
Loading 0:  45%|████▍     | 130/291 [00:33<00:48,  3.34it/s]
Loading 0:  45%|████▌     | 131/291 [00:34<01:00,  2.65it/s]
Loading 0:  46%|████▌     | 134/291 [00:34<00:34,  4.50it/s]
Loading 0:  46%|████▋     | 135/291 [00:34<00:33,  4.71it/s]
Loading 0:  47%|████▋     | 136/291 [00:34<00:29,  5.29it/s]
Loading 0:  47%|████▋     | 138/291 [00:35<00:34,  4.41it/s]
Loading 0:  48%|████▊     | 139/291 [00:35<00:45,  3.33it/s]
Loading 0:  48%|████▊     | 140/291 [00:36<00:57,  2.62it/s]
Loading 0:  49%|████▉     | 143/291 [00:36<00:32,  4.50it/s]
Loading 0:  49%|████▉     | 144/291 [00:36<00:31,  4.74it/s]
Loading 0:  50%|████▉     | 145/291 [00:36<00:27,  5.29it/s]
Loading 0:  51%|█████     | 147/291 [00:37<00:32,  4.41it/s]
Loading 0:  51%|█████     | 148/291 [00:37<00:43,  3.32it/s]
Loading 0:  51%|█████     | 149/291 [00:38<00:54,  2.61it/s]
Loading 0:  52%|█████▏    | 152/291 [00:38<00:31,  4.44it/s]
Loading 0:  53%|█████▎    | 153/291 [00:38<00:29,  4.68it/s]
Loading 0:  53%|█████▎    | 154/291 [00:39<00:26,  5.27it/s]
Loading 0:  54%|█████▎    | 156/291 [00:39<00:30,  4.40it/s]
Loading 0:  54%|█████▍    | 157/291 [00:40<00:40,  3.31it/s]
Loading 0:  54%|█████▍    | 158/291 [00:40<00:50,  2.63it/s]
Loading 0:  55%|█████▌    | 161/291 [00:41<00:29,  4.46it/s]
Loading 0:  56%|█████▌    | 162/291 [00:41<00:27,  4.70it/s]
Loading 0:  56%|█████▌    | 163/291 [00:41<00:24,  5.25it/s]
Loading 0:  57%|█████▋    | 165/291 [00:41<00:28,  4.39it/s]
Loading 0:  57%|█████▋    | 166/291 [00:42<00:37,  3.31it/s]
Loading 0:  57%|█████▋    | 167/291 [00:43<00:47,  2.63it/s]
Loading 0:  58%|█████▊    | 170/291 [00:43<00:27,  4.48it/s]
Loading 0:  59%|█████▉    | 171/291 [00:43<00:25,  4.72it/s]
Loading 0:  59%|█████▉    | 173/291 [00:44<00:30,  3.91it/s]
Loading 0:  60%|██████    | 175/291 [00:44<00:23,  4.91it/s]
Loading 0:  60%|██████    | 176/291 [00:44<00:22,  5.11it/s]
Loading 0:  61%|██████    | 177/291 [00:44<00:20,  5.65it/s]
Loading 0:  62%|██████▏   | 179/291 [00:45<00:24,  4.56it/s]
Loading 0:  62%|██████▏   | 180/291 [00:45<00:32,  3.39it/s]
Loading 0:  62%|██████▏   | 181/291 [00:46<00:41,  2.67it/s]
Loading 0:  63%|██████▎   | 184/291 [00:46<00:23,  4.60it/s]
Loading 0:  64%|██████▎   | 185/291 [00:46<00:21,  4.90it/s]
Loading 0:  64%|██████▍   | 187/291 [00:47<00:18,  5.72it/s]
Loading 0:  65%|██████▍   | 188/291 [00:47<00:26,  3.92it/s]
Loading 0:  65%|██████▍   | 189/291 [00:48<00:34,  2.93it/s]
Loading 0:  66%|██████▌   | 192/291 [00:48<00:26,  3.73it/s]
Loading 0:  66%|██████▋   | 193/291 [00:49<00:31,  3.06it/s]
Loading 0:  67%|██████▋   | 194/291 [00:50<00:38,  2.55it/s]
Loading 0:  68%|██████▊   | 197/291 [00:50<00:22,  4.26it/s]
Loading 0:  68%|██████▊   | 198/291 [00:50<00:20,  4.51it/s]
Loading 0:  69%|██████▉   | 201/291 [00:51<00:20,  4.48it/s]
Loading 0:  69%|██████▉   | 202/291 [00:51<00:25,  3.52it/s]
Loading 0:  70%|██████▉   | 203/291 [00:52<00:31,  2.83it/s]
Loading 0:  71%|███████   | 206/291 [00:52<00:18,  4.56it/s]
Loading 0:  71%|███████   | 207/291 [00:52<00:17,  4.77it/s]
Loading 0:  72%|███████▏  | 210/291 [00:53<00:17,  4.63it/s]
Loading 0:  73%|███████▎  | 211/291 [00:53<00:22,  3.60it/s]
Loading 0:  73%|███████▎  | 212/291 [00:54<00:27,  2.85it/s]
Loading 0:  74%|███████▍  | 215/291 [00:54<00:16,  4.52it/s]
Loading 0:  74%|███████▍  | 216/291 [00:54<00:15,  4.74it/s]
Loading 0:  75%|███████▍  | 217/291 [00:54<00:14,  5.26it/s]
Loading 0:  75%|███████▌  | 219/291 [00:55<00:16,  4.42it/s]
Loading 0:  76%|███████▌  | 220/291 [00:56<00:21,  3.35it/s]
Loading 0:  76%|███████▌  | 221/291 [00:56<00:26,  2.63it/s]
Loading 0:  77%|███████▋  | 224/291 [00:56<00:15,  4.45it/s]
Loading 0:  77%|███████▋  | 225/291 [00:57<00:14,  4.69it/s]
Loading 0:  78%|███████▊  | 226/291 [00:57<00:12,  5.26it/s]
Loading 0:  78%|███████▊  | 228/291 [00:57<00:14,  4.39it/s]
Loading 0:  79%|███████▊  | 229/291 [00:58<00:18,  3.32it/s]
Loading 0:  79%|███████▉  | 230/291 [00:59<00:23,  2.62it/s]
Loading 0:  80%|████████  | 233/291 [00:59<00:13,  4.45it/s]
Loading 0:  80%|████████  | 234/291 [00:59<00:12,  4.69it/s]
Loading 0:  81%|████████▏ | 237/291 [01:00<00:11,  4.58it/s]
Loading 0:  82%|████████▏ | 238/291 [01:00<00:14,  3.55it/s]
Loading 0:  82%|████████▏ | 239/291 [01:01<00:18,  2.84it/s]
Loading 0:  83%|████████▎ | 242/291 [01:01<00:10,  4.61it/s]
Loading 0:  84%|████████▎ | 243/291 [01:01<00:09,  4.83it/s]
Loading 0:  85%|████████▍ | 246/291 [01:02<00:09,  4.68it/s]
Loading 0:  85%|████████▍ | 247/291 [01:02<00:12,  3.65it/s]
Loading 0:  85%|████████▌ | 248/291 [01:03<00:14,  2.91it/s]
Loading 0:  86%|████████▋ | 251/291 [01:03<00:08,  4.66it/s]
Loading 0:  87%|████████▋ | 252/291 [01:03<00:07,  4.94it/s]
Loading 0:  87%|████████▋ | 253/291 [01:03<00:06,  5.47it/s]
Loading 0:  88%|████████▊ | 255/291 [01:04<00:07,  4.57it/s]
Loading 0:  88%|████████▊ | 256/291 [01:05<00:10,  3.44it/s]
Loading 0:  88%|████████▊ | 257/291 [01:05<00:12,  2.69it/s]
Loading 0:  89%|████████▉ | 260/291 [01:05<00:06,  4.54it/s]
Loading 0:  90%|████████▉ | 261/291 [01:06<00:06,  4.77it/s]
Loading 0:  90%|█████████ | 262/291 [01:06<00:05,  5.35it/s]
Loading 0:  91%|█████████ | 264/291 [01:06<00:06,  4.41it/s]
Loading 0:  91%|█████████ | 265/291 [01:07<00:07,  3.33it/s]
Loading 0:  91%|█████████▏| 266/291 [01:07<00:09,  2.64it/s]
Loading 0:  92%|█████████▏| 269/291 [01:08<00:04,  4.54it/s]
Loading 0:  93%|█████████▎| 270/291 [01:08<00:04,  4.78it/s]
Loading 0:  93%|█████████▎| 271/291 [01:08<00:03,  5.35it/s]
Loading 0:  94%|█████████▍| 273/291 [01:08<00:04,  4.49it/s]
Loading 0:  94%|█████████▍| 274/291 [01:09<00:05,  3.39it/s]
Loading 0:  95%|█████████▍| 275/291 [01:10<00:05,  2.67it/s]
Loading 0:  96%|█████████▌| 278/291 [01:10<00:02,  4.58it/s]
Loading 0:  96%|█████████▌| 279/291 [01:10<00:02,  4.82it/s]
Loading 0:  96%|█████████▌| 280/291 [01:10<00:02,  5.31it/s]
Loading 0:  97%|█████████▋| 281/291 [01:11<00:02,  3.62it/s]
Loading 0:  97%|█████████▋| 282/291 [01:11<00:03,  2.75it/s]
Loading 0:  98%|█████████▊| 284/291 [01:12<00:01,  3.92it/s]
Loading 0:  98%|█████████▊| 285/291 [01:12<00:01,  4.27it/s]
Loading 0:  99%|█████████▊| 287/291 [01:12<00:00,  5.28it/s]
Loading 0:  99%|█████████▉| 288/291 [01:13<00:00,  3.63it/s]
                                                            
Job rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer completed after 148.09s with status: succeeded
Stopping job with name rirv938-llama-8b-dpo-vs-2870-v1-mkmlizer
Pipeline stage MKMLizer completed in 153.52s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rirv938-llama-8b-dpo-vs-2870-v1
Waiting for inference service rirv938-llama-8b-dpo-vs-2870-v1 to be ready
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service rirv938-llama-8b-dpo-vs-2870-v1 ready after 160.8956069946289s
Pipeline stage MKMLDeployer completed in 162.61s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 4.915851593017578s
Received healthy response to inference request in 4.328109502792358s
Received healthy response to inference request in 4.77084755897522s
Received healthy response to inference request in 4.705780982971191s
Received healthy response to inference request in 2.4465172290802s
5 requests
0 failed requests
5th percentile: 2.822835683822632
10th percentile: 3.1991541385650635
20th percentile: 3.9517910480499268
30th percentile: 4.403643798828125
40th percentile: 4.554712390899658
50th percentile: 4.705780982971191
60th percentile: 4.731807613372803
70th percentile: 4.757834243774414
80th percentile: 4.799848365783691
90th percentile: 4.857849979400635
95th percentile: 4.886850786209107
99th percentile: 4.910051431655884
mean time: 4.233421373367309
Pipeline stage StressChecker completed in 35.75s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 5.57s
Shutdown handler de-registered
rirv938-llama-8b-dpo-vs-_2870_v1 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Skipping teardown as no inference service was successfully deployed
Pipeline stage MKMLProfilerDeleter completed in 0.13s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.13s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rirv938-llama-8b-dpo-vs-2870-v1-profiler
Waiting for inference service rirv938-llama-8b-dpo-vs-2870-v1-profiler to be ready
Inference service rirv938-llama-8b-dpo-vs-2870-v1-profiler ready after 160.37701559066772s
Pipeline stage MKMLProfilerDeployer completed in 160.79s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rirv938-llama-8b-dpo930240efa1aea95e93e455d45b9e76fa-deplo7rw8s:/code/chaiverse_profiler_1726336800 --namespace tenant-chaiml-guanaco
kubectl exec -it rirv938-llama-8b-dpo930240efa1aea95e93e455d45b9e76fa-deplo7rw8s --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726336800 && python profiles.py profile --best_of_n 1 --auto_batch 5 --batches 1,5,10,15,20,25,30,35,40,45,50,55,60,65,70,75,80,85,90,95,100,105,110,115,120,125,130,135,140,145,150,155,160,165,170,175,180,185,190,195 --samples 200 --input_tokens 256 --output_tokens 1 --summary /code/chaiverse_profiler_1726336800/summary.json'
kubectl exec -it rirv938-llama-8b-dpo930240efa1aea95e93e455d45b9e76fa-deplo7rw8s --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726336800/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1875.83s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rirv938-llama-8b-dpo-vs-2870-v1-profiler is running
Tearing down inference service rirv938-llama-8b-dpo-vs-2870-v1-profiler
Service rirv938-llama-8b-dpo-vs-2870-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.05s
Shutdown handler de-registered
rirv938-llama-8b-dpo-vs-_2870_v1 status is now inactive due to auto deactivation removed underperforming models
rirv938-llama-8b-dpo-vs-_2870_v1 status is now torndown due to DeploymentManager action