developer_uid: Trace2333
submission_id: trace2333-mistral-dpo-trail2_v2
model_name: trace2333-mistral-dpo-trail2_v2
model_group: Trace2333/mistral_dpo_tr
status: torndown
timestamp: 2024-09-11T04:25:50+00:00
num_battles: 11365
num_wins: 5936
celo_rating: 1256.91
family_friendly_score: 0.0
submission_type: basic
model_repo: Trace2333/mistral_dpo_trail2
model_architecture: MistralForCausalLM
model_num_parameters: 12772070400.0
best_of: 8
max_input_tokens: 512
max_output_tokens: 64
latencies: [{'batch_size': 1, 'throughput': 0.6984314751525014, 'latency_mean': 1.4317188358306885, 'latency_p50': 1.4228030443191528, 'latency_p90': 1.5890683650970459}, {'batch_size': 3, 'throughput': 1.3405893508414568, 'latency_mean': 2.236640381813049, 'latency_p50': 2.232417941093445, 'latency_p90': 2.489452362060547}, {'batch_size': 5, 'throughput': 1.5780122066349465, 'latency_mean': 3.1569312870502473, 'latency_p50': 3.16033935546875, 'latency_p90': 3.5947197675704956}, {'batch_size': 6, 'throughput': 1.6224591236275074, 'latency_mean': 3.6621021270751952, 'latency_p50': 3.709444284439087, 'latency_p90': 4.188963484764099}, {'batch_size': 8, 'throughput': 1.6150945359930682, 'latency_mean': 4.9280078411102295, 'latency_p50': 4.956084370613098, 'latency_p90': 5.624430179595947}, {'batch_size': 10, 'throughput': 1.553349684406683, 'latency_mean': 6.387930343151092, 'latency_p50': 6.4569796323776245, 'latency_p90': 7.308911848068237}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: trace2333-mistral-dpo-trail2_v2
is_internal_developer: False
language_model: Trace2333/mistral_dpo_trail2
model_size: 13B
ranking_group: single
throughput_3p7s: 1.64
us_pacific_date: 2024-09-10
win_ratio: 0.5223053233611966
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.06, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '</s>'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}
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}
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
Starting job with name trace2333-mistral-dpo-trail2-v2-mkmlizer
Waiting for job on trace2333-mistral-dpo-trail2-v2-mkmlizer to finish
trace2333-mistral-dpo-trail2-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║     _____            __           __                                ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║       /___/                                                         ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║                                                                     ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║  Version: 0.10.1                                                    ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║  https://mk1.ai                                                     ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║                                                                     ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║  belonging to:                                                      ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║                                                                     ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║  Chai Research Corp.                                                ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ║                                                                     ║
trace2333-mistral-dpo-trail2-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Connection pool is full, discarding connection: %s. Connection pool size: %s
trace2333-mistral-dpo-trail2-v2-mkmlizer: Downloaded to shared memory in 29.191s
trace2333-mistral-dpo-trail2-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpdmc797u4, device:0
trace2333-mistral-dpo-trail2-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
trace2333-mistral-dpo-trail2-v2-mkmlizer: quantized model in 34.494s
trace2333-mistral-dpo-trail2-v2-mkmlizer: Processed model Trace2333/mistral_dpo_trail2 in 63.685s
trace2333-mistral-dpo-trail2-v2-mkmlizer: creating bucket guanaco-mkml-models
trace2333-mistral-dpo-trail2-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
trace2333-mistral-dpo-trail2-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/trace2333-mistral-dpo-trail2-v2
trace2333-mistral-dpo-trail2-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/trace2333-mistral-dpo-trail2-v2/config.json
trace2333-mistral-dpo-trail2-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/trace2333-mistral-dpo-trail2-v2/tokenizer.json
trace2333-mistral-dpo-trail2-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/trace2333-mistral-dpo-trail2-v2/flywheel_model.0.safetensors
trace2333-mistral-dpo-trail2-v2-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/363 [00:00<00:06, 56.69it/s]
Loading 0:   6%|▌         | 22/363 [00:00<00:03, 94.79it/s]
Loading 0:   9%|▉         | 34/363 [00:00<00:03, 88.15it/s]
Loading 0:  12%|█▏        | 44/363 [00:00<00:03, 91.55it/s]
Loading 0:  16%|█▌        | 58/363 [00:00<00:03, 99.02it/s]
Loading 0:  19%|█▊        | 68/363 [00:01<00:11, 25.32it/s]
Loading 0:  21%|██        | 76/363 [00:01<00:09, 30.58it/s]
Loading 0:  24%|██▍       | 88/363 [00:01<00:06, 39.88it/s]
Loading 0:  28%|██▊       | 103/363 [00:02<00:04, 53.35it/s]
Loading 0:  31%|███       | 113/363 [00:02<00:04, 60.21it/s]
Loading 0:  34%|███▍      | 123/363 [00:02<00:03, 67.65it/s]
Loading 0:  37%|███▋      | 133/363 [00:02<00:03, 70.22it/s]
Loading 0:  39%|███▉      | 142/363 [00:03<00:09, 23.59it/s]
Loading 0:  42%|████▏     | 151/363 [00:03<00:07, 29.55it/s]
Loading 0:  46%|████▌     | 166/363 [00:03<00:04, 41.62it/s]
Loading 0:  49%|████▉     | 178/363 [00:03<00:03, 49.99it/s]
Loading 0:  52%|█████▏    | 188/363 [00:03<00:03, 57.80it/s]
Loading 0:  56%|█████▌    | 202/363 [00:04<00:02, 68.54it/s]
Loading 0:  59%|█████▉    | 214/363 [00:04<00:02, 73.76it/s]
Loading 0:  62%|██████▏   | 224/363 [00:05<00:05, 25.62it/s]
Loading 0:  66%|██████▌   | 238/363 [00:05<00:03, 34.63it/s]
Loading 0:  69%|██████▉   | 250/363 [00:05<00:02, 42.40it/s]
Loading 0:  73%|███████▎  | 265/363 [00:05<00:01, 53.90it/s]
Loading 0:  76%|███████▌  | 275/363 [00:05<00:01, 60.51it/s]
Loading 0:  79%|███████▉  | 286/363 [00:05<00:01, 64.98it/s]
Loading 0:  83%|████████▎ | 301/363 [00:06<00:00, 75.59it/s]
Loading 0:  86%|████████▌ | 311/363 [00:07<00:01, 27.09it/s]
Loading 0:  88%|████████▊ | 321/363 [00:07<00:01, 33.56it/s]
Loading 0:  91%|█████████ | 331/363 [00:07<00:00, 39.03it/s]
Loading 0:  94%|█████████▍| 341/363 [00:07<00:00, 47.04it/s]
Loading 0:  98%|█████████▊| 355/363 [00:07<00:00, 59.15it/s]
                                                            

Job trace2333-mistral-dpo-trail2-v2-mkmlizer completed after 84.72s with status: succeeded
Stopping job with name trace2333-mistral-dpo-trail2-v2-mkmlizer
Pipeline stage MKMLizer completed in 85.61s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service trace2333-mistral-dpo-trail2-v2
Waiting for inference service trace2333-mistral-dpo-trail2-v2 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
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
Inference service trace2333-mistral-dpo-trail2-v2 ready after 171.70197105407715s
Pipeline stage MKMLDeployer completed in 172.52s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.389510154724121s
Received healthy response to inference request in 2.000657558441162s
Received healthy response to inference request in 2.0514185428619385s
Received healthy response to inference request in 3.0342679023742676s
Received healthy response to inference request in 1.9494690895080566s
5 requests
0 failed requests
5th percentile: 1.9597067832946777
10th percentile: 1.9699444770812988
20th percentile: 1.990419864654541
30th percentile: 2.0108097553253175
40th percentile: 2.0311141490936278
50th percentile: 2.0514185428619385
60th percentile: 2.1866551876068114
70th percentile: 2.3218918323516844
80th percentile: 2.5184617042541504
90th percentile: 2.776364803314209
95th percentile: 2.9053163528442383
99th percentile: 3.0084775924682616
mean time: 2.2850646495819094
Pipeline stage StressChecker completed in 13.42s
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 8.86s
Shutdown handler de-registered
trace2333-mistral-dpo-trail2_v2 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 0.12s
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 trace2333-mistral-dpo-trail2-v2-profiler
Waiting for inference service trace2333-mistral-dpo-trail2-v2-profiler to be ready
Inference service trace2333-mistral-dpo-trail2-v2-profiler ready after 160.3755795955658s
Pipeline stage MKMLProfilerDeployer completed in 160.73s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/trace2333-mistral-dp24be3044b00bbd44188ef567de61e945-deplo9k84h:/code/chaiverse_profiler_1726029234 --namespace tenant-chaiml-guanaco
kubectl exec -it trace2333-mistral-dp24be3044b00bbd44188ef567de61e945-deplo9k84h --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726029234 && python profiles.py profile --best_of_n 8 --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 512 --output_tokens 64 --summary /code/chaiverse_profiler_1726029234/summary.json'
kubectl exec -it trace2333-mistral-dp24be3044b00bbd44188ef567de61e945-deplo9k84h --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726029234/summary.json'
Pipeline stage MKMLProfilerRunner completed in 943.95s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service trace2333-mistral-dpo-trail2-v2-profiler is running
Tearing down inference service trace2333-mistral-dpo-trail2-v2-profiler
Service trace2333-mistral-dpo-trail2-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.84s
Shutdown handler de-registered
trace2333-mistral-dpo-trail2_v2 status is now inactive due to auto deactivation removed underperforming models
trace2333-mistral-dpo-trail2_v2 status is now torndown due to DeploymentManager action