Running pipeline stage MKMLizer
Starting job with name trace2333-fd-llama3-v4-v2-mkmlizer
Waiting for job on trace2333-fd-llama3-v4-v2-mkmlizer to finish
trace2333-fd-llama3-v4-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
trace2333-fd-llama3-v4-v2-mkmlizer: ║ _____ __ __ ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ /___/ ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ Version: 0.9.9 ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ https://mk1.ai ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ The license key for the current software has been verified as ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ belonging to: ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ Chai Research Corp. ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
trace2333-fd-llama3-v4-v2-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
trace2333-fd-llama3-v4-v2-mkmlizer: Downloaded to shared memory in 60.098s
trace2333-fd-llama3-v4-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp0596cdgf, device:0
trace2333-fd-llama3-v4-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
trace2333-fd-llama3-v4-v2-mkmlizer: quantized model in 29.308s
trace2333-fd-llama3-v4-v2-mkmlizer: Processed model Trace2333/fd_llama3_v4 in 89.406s
trace2333-fd-llama3-v4-v2-mkmlizer: creating bucket guanaco-mkml-models
trace2333-fd-llama3-v4-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
trace2333-fd-llama3-v4-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v2
trace2333-fd-llama3-v4-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v2/config.json
trace2333-fd-llama3-v4-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v2/tokenizer_config.json
trace2333-fd-llama3-v4-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v2/special_tokens_map.json
trace2333-fd-llama3-v4-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v2/tokenizer.json
trace2333-fd-llama3-v4-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v2/flywheel_model.0.safetensors
trace2333-fd-llama3-v4-v2-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.
trace2333-fd-llama3-v4-v2-mkmlizer: warnings.warn(
trace2333-fd-llama3-v4-v2-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.
trace2333-fd-llama3-v4-v2-mkmlizer: warnings.warn(
trace2333-fd-llama3-v4-v2-mkmlizer: Saving duration: 1.408s
trace2333-fd-llama3-v4-v2-mkmlizer: Processed model Jellywibble/gpt2_xl_pairwise_89m_step_347634 in 11.670s
trace2333-fd-llama3-v4-v2-mkmlizer: creating bucket guanaco-reward-models
trace2333-fd-llama3-v4-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
trace2333-fd-llama3-v4-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/trace2333-fd-llama3-v4-v2_reward
trace2333-fd-llama3-v4-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v2_reward/config.json
trace2333-fd-llama3-v4-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v2_reward/special_tokens_map.json
trace2333-fd-llama3-v4-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v2_reward/tokenizer_config.json
trace2333-fd-llama3-v4-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/trace2333-fd-llama3-v4-v2_reward/merges.txt
trace2333-fd-llama3-v4-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v2_reward/vocab.json
trace2333-fd-llama3-v4-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v2_reward/tokenizer.json
trace2333-fd-llama3-v4-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/trace2333-fd-llama3-v4-v2_reward/reward.tensors
Job trace2333-fd-llama3-v4-v2-mkmlizer completed after 136.14s with status: succeeded
Stopping job with name trace2333-fd-llama3-v4-v2-mkmlizer
Pipeline stage MKMLizer completed in 137.26s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.08s
Running pipeline stage ISVCDeployer
Creating inference service trace2333-fd-llama3-v4-v2
Waiting for inference service trace2333-fd-llama3-v4-v2 to be ready
Inference service trace2333-fd-llama3-v4-v2 ready after 160.9782145023346s
Pipeline stage ISVCDeployer completed in 162.28s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.33211350440979s
Received healthy response to inference request in 1.5092244148254395s
Received healthy response to inference request in 1.468503713607788s
Received healthy response to inference request in 1.4006764888763428s
Received healthy response to inference request in 1.3820888996124268s
5 requests
0 failed requests
5th percentile: 1.38580641746521
10th percentile: 1.3895239353179931
20th percentile: 1.3969589710235595
30th percentile: 1.4142419338226317
40th percentile: 1.44137282371521
50th percentile: 1.468503713607788
60th percentile: 1.4847919940948486
70th percentile: 1.5010802745819092
80th percentile: 1.6738022327423097
90th percentile: 2.00295786857605
95th percentile: 2.16753568649292
99th percentile: 2.299197940826416
mean time: 1.6185214042663574
Pipeline stage StressChecker completed in 8.73s
trace2333-fd-llama3-v4_v2 status is now deployed due to DeploymentManager action
trace2333-fd-llama3-v4_v2 status is now inactive due to auto deactivation removed underperforming models
trace2333-fd-llama3-v4_v2 status is now torndown due to DeploymentManager action