Running pipeline stage MKMLizer
Starting job with name pjmixers-llama-3-curseds-4945-v1-mkmlizer
Waiting for job on pjmixers-llama-3-curseds-4945-v1-mkmlizer to finish
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ _____ __ __ ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ /___/ ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ Version: 0.9.7 ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ https://mk1.ai ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ The license key for the current software has been verified as ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ belonging to: ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ Chai Research Corp. ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ║ ║
pjmixers-llama-3-curseds-4945-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
pjmixers-llama-3-curseds-4945-v1-mkmlizer: Downloaded to shared memory in 63.242s
pjmixers-llama-3-curseds-4945-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpk_2b0y0d, device:0
pjmixers-llama-3-curseds-4945-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
pjmixers-llama-3-curseds-4945-v1-mkmlizer: quantized model in 28.195s
pjmixers-llama-3-curseds-4945-v1-mkmlizer: Processed model PJMixers/LLaMa-3-CursedStock-v2.0-8B in 91.437s
pjmixers-llama-3-curseds-4945-v1-mkmlizer: creating bucket guanaco-mkml-models
pjmixers-llama-3-curseds-4945-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
pjmixers-llama-3-curseds-4945-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/pjmixers-llama-3-curseds-4945-v1
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/pjmixers-llama-3-curseds-4945-v1/config.json
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/pjmixers-llama-3-curseds-4945-v1/special_tokens_map.json
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/pjmixers-llama-3-curseds-4945-v1/tokenizer_config.json
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/pjmixers-llama-3-curseds-4945-v1/tokenizer.json
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/pjmixers-llama-3-curseds-4945-v1/flywheel_model.0.safetensors
pjmixers-llama-3-curseds-4945-v1-mkmlizer: loading reward model from Jellywibble/gpt2_xl_pairwise_89m_step_347634
pjmixers-llama-3-curseds-4945-v1-mkmlizer:
Loading 0: 0%| | 0/291 [00:00<?, ?it/s]
Loading 0: 1%| | 2/291 [00:04<11:26, 2.37s/it]
Loading 0: 2%|▏ | 6/291 [00:04<03:05, 1.54it/s]
Loading 0: 3%|▎ | 10/291 [00:05<01:37, 2.87it/s]
Loading 0: 5%|▍ | 14/291 [00:05<00:59, 4.67it/s]
Loading 0: 7%|▋ | 19/291 [00:05<00:35, 7.62it/s]
Loading 0: 8%|▊ | 23/291 [00:05<00:26, 10.17it/s]
Loading 0: 9%|▉ | 27/291 [00:05<00:19, 13.29it/s]
Loading 0: 11%|█ | 31/291 [00:05<00:15, 16.80it/s]
Loading 0: 12%|█▏ | 35/291 [00:05<00:12, 19.80it/s]
Loading 0: 14%|█▎ | 40/291 [00:05<00:10, 24.87it/s]
Loading 0: 15%|█▌ | 44/291 [00:06<00:09, 26.78it/s]
Loading 0: 17%|█▋ | 49/291 [00:06<00:07, 31.31it/s]
Loading 0: 18%|█▊ | 53/291 [00:06<00:07, 31.24it/s]
Loading 0: 20%|█▉ | 58/291 [00:06<00:06, 35.30it/s]
Loading 0: 22%|██▏ | 63/291 [00:06<00:09, 25.22it/s]
Loading 0: 23%|██▎ | 67/291 [00:06<00:08, 27.88it/s]
Loading 0: 24%|██▍ | 71/291 [00:06<00:07, 28.47it/s]
Loading 0: 26%|██▌ | 76/291 [00:07<00:06, 32.44it/s]
Loading 0: 27%|██▋ | 80/291 [00:07<00:06, 32.24it/s]
Loading 0: 29%|██▉ | 85/291 [00:07<00:05, 35.91it/s]
Loading 0: 31%|███ | 89/291 [00:07<00:05, 34.58it/s]
Loading 0: 32%|███▏ | 94/291 [00:07<00:05, 36.63it/s]
Loading 0: 34%|███▎ | 98/291 [00:07<00:05, 34.98it/s]
Loading 0: 35%|███▌ | 103/291 [00:07<00:05, 37.09it/s]
Loading 0: 37%|███▋ | 107/291 [00:07<00:05, 35.40it/s]
Loading 0: 38%|███▊ | 112/291 [00:08<00:04, 38.05it/s]
Loading 0: 40%|███▉ | 116/291 [00:08<00:07, 24.12it/s]
Loading 0: 42%|████▏ | 121/291 [00:08<00:06, 28.07it/s]
Loading 0: 43%|████▎ | 125/291 [00:08<00:05, 28.71it/s]
Loading 0: 45%|████▍ | 130/291 [00:08<00:05, 32.19it/s]
Loading 0: 46%|████▌ | 134/291 [00:08<00:04, 32.14it/s]
Loading 0: 48%|████▊ | 139/291 [00:08<00:04, 35.85it/s]
Loading 0: 49%|████▉ | 143/291 [00:09<00:04, 34.93it/s]
Loading 0: 51%|█████ | 148/291 [00:09<00:03, 36.38it/s]
Loading 0: 52%|█████▏ | 152/291 [00:09<00:04, 34.45it/s]
Loading 0: 54%|█████▍ | 157/291 [00:09<00:03, 36.87it/s]
Loading 0: 55%|█████▌ | 161/291 [00:09<00:03, 34.99it/s]
Loading 0: 57%|█████▋ | 166/291 [00:09<00:04, 26.92it/s]
Loading 0: 58%|█████▊ | 170/291 [00:10<00:04, 28.05it/s]
Loading 0: 60%|██████ | 175/291 [00:10<00:03, 31.71it/s]
Loading 0: 62%|██████▏ | 179/291 [00:10<00:03, 31.73it/s]
Loading 0: 63%|██████▎ | 184/291 [00:10<00:03, 35.19it/s]
Loading 0: 65%|██████▍ | 188/291 [00:10<00:03, 32.91it/s]
Loading 0: 66%|██████▋ | 193/291 [00:10<00:02, 35.35it/s]
Loading 0: 68%|██████▊ | 197/291 [00:10<00:02, 33.85it/s]
Loading 0: 69%|██████▉ | 202/291 [00:10<00:02, 36.88it/s]
Loading 0: 71%|███████ | 206/291 [00:11<00:02, 35.47it/s]
Loading 0: 73%|███████▎ | 211/291 [00:11<00:02, 37.99it/s]
Loading 0: 74%|███████▍ | 215/291 [00:11<00:03, 24.67it/s]
Loading 0: 76%|███████▌ | 220/291 [00:11<00:02, 29.00it/s]
Loading 0: 77%|███████▋ | 224/291 [00:11<00:02, 29.62it/s]
Loading 0: 79%|███████▊ | 229/291 [00:11<00:01, 32.98it/s]
Loading 0: 80%|████████ | 233/291 [00:11<00:01, 32.39it/s]
Loading 0: 82%|████████▏ | 238/291 [00:12<00:01, 34.17it/s]
Loading 0: 83%|████████▎ | 242/291 [00:12<00:01, 33.48it/s]
Loading 0: 85%|████████▍ | 247/291 [00:12<00:01, 36.39it/s]
Loading 0: 86%|████████▋ | 251/291 [00:12<00:01, 35.07it/s]
Loading 0: 88%|████████▊ | 256/291 [00:12<00:00, 37.73it/s]
Loading 0: 89%|████████▉ | 260/291 [00:12<00:00, 36.16it/s]
Loading 0: 91%|█████████ | 265/291 [00:12<00:00, 38.27it/s]
Loading 0: 92%|█████████▏| 269/291 [00:13<00:00, 25.03it/s]
Loading 0: 94%|█████████▍| 274/291 [00:13<00:00, 29.27it/s]
Loading 0: 96%|█████████▌| 278/291 [00:13<00:00, 29.99it/s]
Loading 0: 97%|█████████▋| 283/291 [00:13<00:00, 33.70it/s]
Loading 0: 99%|█████████▊| 287/291 [00:13<00:00, 33.18it/s]
/opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:957: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
pjmixers-llama-3-curseds-4945-v1-mkmlizer: warnings.warn(
pjmixers-llama-3-curseds-4945-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:785: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
pjmixers-llama-3-curseds-4945-v1-mkmlizer: warnings.warn(
pjmixers-llama-3-curseds-4945-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:469: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
pjmixers-llama-3-curseds-4945-v1-mkmlizer: warnings.warn(
pjmixers-llama-3-curseds-4945-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
pjmixers-llama-3-curseds-4945-v1-mkmlizer: Saving duration: 1.377s
pjmixers-llama-3-curseds-4945-v1-mkmlizer: Processed model Jellywibble/gpt2_xl_pairwise_89m_step_347634 in 10.623s
pjmixers-llama-3-curseds-4945-v1-mkmlizer: creating bucket guanaco-reward-models
pjmixers-llama-3-curseds-4945-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
pjmixers-llama-3-curseds-4945-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/pjmixers-llama-3-curseds-4945-v1_reward
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/pjmixers-llama-3-curseds-4945-v1_reward/config.json
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/pjmixers-llama-3-curseds-4945-v1_reward/special_tokens_map.json
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/pjmixers-llama-3-curseds-4945-v1_reward/tokenizer_config.json
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/pjmixers-llama-3-curseds-4945-v1_reward/merges.txt
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/pjmixers-llama-3-curseds-4945-v1_reward/vocab.json
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/pjmixers-llama-3-curseds-4945-v1_reward/tokenizer.json
pjmixers-llama-3-curseds-4945-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/pjmixers-llama-3-curseds-4945-v1_reward/reward.tensors
Job pjmixers-llama-3-curseds-4945-v1-mkmlizer completed after 136.46s with status: succeeded
Stopping job with name pjmixers-llama-3-curseds-4945-v1-mkmlizer
Pipeline stage MKMLizer completed in 137.48s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.12s
Running pipeline stage ISVCDeployer
Creating inference service pjmixers-llama-3-curseds-4945-v1
Waiting for inference service pjmixers-llama-3-curseds-4945-v1 to be ready
Inference service pjmixers-llama-3-curseds-4945-v1 ready after 80.69404888153076s
Pipeline stage ISVCDeployer completed in 82.28s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1022884845733643s
Received healthy response to inference request in 1.3339664936065674s
Received healthy response to inference request in 1.3577723503112793s
Received healthy response to inference request in 1.3706471920013428s
Received healthy response to inference request in 1.3587901592254639s
5 requests
0 failed requests
5th percentile: 1.3387276649475097
10th percentile: 1.343488836288452
20th percentile: 1.353011178970337
30th percentile: 1.3579759120941162
40th percentile: 1.3583830356597901
50th percentile: 1.3587901592254639
60th percentile: 1.3635329723358154
70th percentile: 1.368275785446167
80th percentile: 1.5169754505157471
90th percentile: 1.8096319675445558
95th percentile: 1.95596022605896
99th percentile: 2.073022832870483
mean time: 1.5046929359436034
Pipeline stage StressChecker completed in 8.23s
pjmixers-llama-3-curseds_4945_v1 status is now deployed due to DeploymentManager action
pjmixers-llama-3-curseds_4945_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of pjmixers-llama-3-curseds_4945_v1
Running pipeline stage ISVCDeleter
Checking if service pjmixers-llama-3-curseds-4945-v1 is running
Tearing down inference service pjmixers-llama-3-curseds-4945-v1
Service pjmixers-llama-3-curseds-4945-v1 has been torndown
Pipeline stage ISVCDeleter completed in 4.82s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key pjmixers-llama-3-curseds-4945-v1/config.json from bucket guanaco-mkml-models
Deleting key pjmixers-llama-3-curseds-4945-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key pjmixers-llama-3-curseds-4945-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key pjmixers-llama-3-curseds-4945-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key pjmixers-llama-3-curseds-4945-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key pjmixers-llama-3-curseds-4945-v1_reward/config.json from bucket guanaco-reward-models
Deleting key pjmixers-llama-3-curseds-4945-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key pjmixers-llama-3-curseds-4945-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key pjmixers-llama-3-curseds-4945-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key pjmixers-llama-3-curseds-4945-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key pjmixers-llama-3-curseds-4945-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key pjmixers-llama-3-curseds-4945-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.49s
pjmixers-llama-3-curseds_4945_v1 status is now torndown due to DeploymentManager action