chaiml-albert-dpo-0912-v_8296

submission_id: chaiml-albert-dpo-0912-v_8296_v3

developer_uid: chai_backend_admin

alignment_samples: 10545

alignment_score: 0.29438774489837594

best_of: 16

celo_rating: 1251.95

display_name: dpo-ai-selected-16-512

formatter: {'memory_template': '', 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}

generation_params: {'temperature': 0.9, 'top_p': 1.0, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '</s>', '###', 'Bot:', 'User:', 'You:', '<|im_end|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

is_internal_developer: True

language_model: ChaiML/albert_dpo_0912_virgo_edit_ai_selected

latencies: [{'batch_size': 1, 'throughput': 0.629650173787253, 'latency_mean': 1.588126196861267, 'latency_p50': 1.5808131694793701, 'latency_p90': 1.7412659406661988}, {'batch_size': 2, 'throughput': 0.8511974195545513, 'latency_mean': 2.3462184262275696, 'latency_p50': 2.3531265258789062, 'latency_p90': 2.6189200401306154}, {'batch_size': 3, 'throughput': 0.933961917777532, 'latency_mean': 3.206421080827713, 'latency_p50': 3.198436737060547, 'latency_p90': 3.5645790338516234}, {'batch_size': 4, 'throughput': 0.933301267835482, 'latency_mean': 4.274065288305283, 'latency_p50': 4.257059216499329, 'latency_p90': 4.878874063491821}, {'batch_size': 5, 'throughput': 0.8908083929893296, 'latency_mean': 5.602509803771973, 'latency_p50': 5.622680425643921, 'latency_p90': 6.285459899902344}]

max_input_tokens: 512

max_output_tokens: 64

model_architecture: MistralForCausalLM

model_group: ChaiML/albert_dpo_0912_v

model_name: dpo-ai-selected-16-512

model_num_parameters: 12772070400.0

model_repo: ChaiML/albert_dpo_0912_virgo_edit_ai_selected

model_size: 13B

num_battles: 10545

num_wins: 5361

propriety_score: 0.7178683385579937

propriety_total_count: 957.0

ranking_group: single

status: inactive

submission_type: basic

throughput_3p7s: 0.94

timestamp: 2024-09-13T20:42:31+00:00

us_pacific_date: 2024-09-13

win_ratio: 0.5083926031294452

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
Starting job with name chaiml-albert-dpo-0912-v-8296-v3-mkmlizer
Waiting for job on chaiml-albert-dpo-0912-v-8296-v3-mkmlizer to finish
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║     _____            __           __                                ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║       /___/                                                         ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║  Version: 0.10.1                                                    ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║  belonging to:                                                      ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: Downloaded to shared memory in 54.902s
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmplzlqtaup, device:0
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: quantized model in 41.981s
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: Processed model ChaiML/albert_dpo_0912_virgo_edit_ai_selected in 96.884s
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: creating bucket guanaco-mkml-models
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v3
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v3/config.json
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v3/special_tokens_map.json
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v3/tokenizer_config.json
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v3/tokenizer.json
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v3/flywheel_model.0.safetensors
chaiml-albert-dpo-0912-v-8296-v3-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 4/363 [00:00<00:09, 39.51it/s]
Loading 0:   2%|▏         | 8/363 [00:00<00:13, 26.33it/s]
Loading 0:   3%|▎         | 12/363 [00:00<00:12, 28.62it/s]
Loading 0:   4%|▍         | 16/363 [00:00<00:12, 27.33it/s]
Loading 0:   6%|▌         | 21/363 [00:00<00:10, 33.26it/s]
Loading 0:   7%|▋         | 25/363 [00:01<00:16, 20.62it/s]
Loading 0:   8%|▊         | 28/363 [00:01<00:15, 21.26it/s]
Loading 0:   9%|▉         | 32/363 [00:01<00:16, 20.68it/s]
Loading 0:  10%|█         | 37/363 [00:01<00:12, 25.73it/s]
Loading 0:  11%|█▏        | 41/363 [00:01<00:13, 23.67it/s]
Loading 0:  13%|█▎        | 46/363 [00:01<00:10, 28.88it/s]
Loading 0:  14%|█▍        | 50/363 [00:01<00:12, 25.18it/s]
Loading 0:  15%|█▌        | 55/363 [00:02<00:10, 29.89it/s]
Loading 0:  17%|█▋        | 60/363 [00:02<00:09, 30.80it/s]
Loading 0:  18%|█▊        | 64/363 [00:02<00:14, 20.45it/s]
Loading 0:  20%|█▉        | 71/363 [00:02<00:11, 26.37it/s]
Loading 0:  21%|██        | 75/363 [00:02<00:10, 26.33it/s]
Loading 0:  22%|██▏       | 79/363 [00:03<00:10, 26.14it/s]
Loading 0:  23%|██▎       | 84/363 [00:03<00:09, 29.18it/s]
Loading 0:  24%|██▍       | 88/363 [00:03<00:10, 27.42it/s]
Loading 0:  26%|██▌       | 93/363 [00:03<00:08, 30.00it/s]
Loading 0:  27%|██▋       | 97/363 [00:03<00:09, 27.97it/s]
Loading 0:  28%|██▊       | 101/363 [00:03<00:11, 23.02it/s]
Loading 0:  29%|██▊       | 104/363 [00:04<00:12, 20.71it/s]
Loading 0:  31%|███       | 111/363 [00:04<00:09, 27.57it/s]
Loading 0:  32%|███▏      | 115/363 [00:04<00:09, 26.96it/s]
Loading 0:  33%|███▎      | 120/363 [00:04<00:08, 29.87it/s]
Loading 0:  34%|███▍      | 124/363 [00:04<00:08, 28.19it/s]
Loading 0:  36%|███▌      | 129/363 [00:04<00:07, 30.99it/s]
Loading 0:  37%|███▋      | 133/363 [00:05<00:07, 29.39it/s]
Loading 0:  38%|███▊      | 137/363 [00:05<00:07, 29.88it/s]
Loading 0:  39%|███▉      | 142/363 [00:05<00:08, 25.46it/s]
Loading 0:  40%|███▉      | 145/363 [00:05<00:09, 23.81it/s]
Loading 0:  41%|████      | 149/363 [00:05<00:09, 22.19it/s]
Loading 0:  42%|████▏     | 154/363 [00:05<00:07, 27.41it/s]
Loading 0:  44%|████▎     | 158/363 [00:06<00:08, 25.35it/s]
Loading 0:  45%|████▍     | 163/363 [00:06<00:06, 30.08it/s]
Loading 0:  46%|████▌     | 167/363 [00:06<00:07, 26.10it/s]
Loading 0:  48%|████▊     | 174/363 [00:06<00:05, 31.89it/s]
Loading 0:  49%|████▉     | 178/363 [00:06<00:06, 29.42it/s]
Loading 0:  50%|█████     | 182/363 [00:06<00:07, 25.05it/s]
Loading 0:  51%|█████     | 185/363 [00:07<00:08, 21.99it/s]
Loading 0:  53%|█████▎    | 192/363 [00:07<00:05, 28.77it/s]
Loading 0:  54%|█████▍    | 196/363 [00:07<00:05, 27.98it/s]
Loading 0:  55%|█████▌    | 201/363 [00:07<00:05, 29.67it/s]
Loading 0:  56%|█████▋    | 205/363 [00:07<00:05, 28.24it/s]
Loading 0:  58%|█████▊    | 210/363 [00:07<00:05, 29.99it/s]
Loading 0:  59%|█████▉    | 214/363 [00:07<00:05, 28.29it/s]
Loading 0:  60%|██████    | 218/363 [00:08<00:05, 28.57it/s]
Loading 0:  61%|██████▏   | 223/363 [00:08<00:05, 25.04it/s]
Loading 0:  62%|██████▏   | 226/363 [00:08<00:05, 23.57it/s]
Loading 0:  63%|██████▎   | 230/363 [00:08<00:06, 21.93it/s]
Loading 0:  65%|██████▍   | 235/363 [00:08<00:04, 26.78it/s]
Loading 0:  66%|██████▌   | 238/363 [00:08<00:04, 27.33it/s]
Loading 0:  66%|██████▋   | 241/363 [00:09<00:04, 26.42it/s]
Loading 0:  68%|██████▊   | 246/363 [00:09<00:04, 28.56it/s]
Loading 0:  69%|██████▊   | 249/363 [00:09<00:04, 25.95it/s]
Loading 0:  70%|███████   | 255/363 [00:09<00:03, 30.39it/s]
Loading 0:  71%|███████▏  | 259/363 [00:09<00:03, 29.30it/s]
Loading 0:  72%|███████▏  | 263/363 [00:09<00:04, 24.15it/s]
Loading 0:  73%|███████▎  | 266/363 [00:10<00:04, 21.00it/s]
Loading 0:  75%|███████▌  | 273/363 [00:10<00:03, 27.80it/s]
Loading 0:  76%|███████▌  | 276/363 [00:10<00:03, 25.40it/s]
Loading 0:  77%|███████▋  | 280/363 [00:10<00:02, 27.81it/s]
Loading 0:  78%|███████▊  | 284/363 [00:10<00:03, 24.78it/s]
Loading 0:  80%|████████  | 291/363 [00:10<00:02, 30.50it/s]
Loading 0:  81%|████████▏ | 295/363 [00:11<00:02, 28.16it/s]
Loading 0:  82%|████████▏ | 299/363 [00:11<00:02, 28.36it/s]
Loading 0:  84%|████████▎ | 304/363 [00:11<00:02, 24.62it/s]
Loading 0:  85%|████████▍ | 307/363 [00:11<00:02, 22.95it/s]
Loading 0:  86%|████████▌ | 311/363 [00:11<00:02, 22.38it/s]
Loading 0:  87%|████████▋ | 316/363 [00:11<00:01, 27.48it/s]
Loading 0:  88%|████████▊ | 320/363 [00:12<00:01, 24.96it/s]
Loading 0:  90%|█████████ | 327/363 [00:12<00:01, 31.77it/s]
Loading 0:  91%|█████████ | 331/363 [00:12<00:01, 30.02it/s]
Loading 0:  93%|█████████▎| 336/363 [00:12<00:00, 31.34it/s]
Loading 0:  94%|█████████▎| 340/363 [00:12<00:00, 29.59it/s]
Loading 0:  95%|█████████▍| 344/363 [00:19<00:09,  1.99it/s]
Loading 0:  96%|█████████▌| 348/363 [00:19<00:05,  2.66it/s]
Loading 0:  97%|█████████▋| 353/363 [00:20<00:02,  3.85it/s]
Loading 0:  98%|█████████▊| 357/363 [00:20<00:01,  4.95it/s]
Loading 0: 100%|█████████▉| 362/363 [00:20<00:00,  7.03it/s]
                                                            
Job chaiml-albert-dpo-0912-v-8296-v3-mkmlizer completed after 117.48s with status: succeeded
Stopping job with name chaiml-albert-dpo-0912-v-8296-v3-mkmlizer
Pipeline stage MKMLizer completed in 118.62s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.11s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-albert-dpo-0912-v-8296-v3
Waiting for inference service chaiml-albert-dpo-0912-v-8296-v3 to be ready
Inference service chaiml-albert-dpo-0912-v-8296-v3 ready after 191.94127678871155s
Pipeline stage MKMLDeployer completed in 192.92s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.5773751735687256s
Received healthy response to inference request in 2.2138442993164062s
Received healthy response to inference request in 2.0677175521850586s
Received healthy response to inference request in 2.2867307662963867s
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
Received healthy response to inference request in 3.102597713470459s
5 requests
0 failed requests
5th percentile: 2.096942901611328
10th percentile: 2.1261682510375977
20th percentile: 2.1846189498901367
30th percentile: 2.2284215927124023
40th percentile: 2.2575761795043947
50th percentile: 2.2867307662963867
60th percentile: 2.4029885292053224
70th percentile: 2.5192462921142575
80th percentile: 2.682419681549072
90th percentile: 2.892508697509766
95th percentile: 2.9975532054901124
99th percentile: 3.0815888118743895
mean time: 2.4496531009674074
Pipeline stage StressChecker completed in 13.54s
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 6.82s
Shutdown handler de-registered
chaiml-albert-dpo-0912-v_8296_v3 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 0.10s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.10s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-albert-dpo-0912-v-8296-v3-profiler
Waiting for inference service chaiml-albert-dpo-0912-v-8296-v3-profiler to be ready
Inference service chaiml-albert-dpo-0912-v-8296-v3-profiler ready after 190.42307567596436s
Pipeline stage MKMLProfilerDeployer completed in 190.79s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-albert-dpo-09ee20ddd26b6a3a5dd7e3b194fba889fe-deplo9zh5l:/code/chaiverse_profiler_1726260714 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-albert-dpo-09ee20ddd26b6a3a5dd7e3b194fba889fe-deplo9zh5l --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726260714 && python profiles.py profile --best_of_n 16 --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_1726260714/summary.json'
kubectl exec -it chaiml-albert-dpo-09ee20ddd26b6a3a5dd7e3b194fba889fe-deplo9zh5l --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726260714/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1210.72s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-albert-dpo-0912-v-8296-v3-profiler is running
Tearing down inference service chaiml-albert-dpo-0912-v-8296-v3-profiler
Service chaiml-albert-dpo-0912-v-8296-v3-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.95s
Shutdown handler de-registered
chaiml-albert-dpo-0912-v_8296_v3 status is now inactive due to auto deactivation removed underperforming models