chaiml-albert-dpo-0912-v_8296

submission_id: chaiml-albert-dpo-0912-v_8296_v2

developer_uid: chai_backend_admin

alignment_samples: 10349

alignment_score: 0.7654179740449221

best_of: 8

celo_rating: 1240.99

display_name: dpo-ai-selected

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': 1024, 'best_of': 8, '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': 0.6098770314778904, 'latency_mean': 1.639603341817856, 'latency_p50': 1.6371160745620728, 'latency_p90': 1.8051660060882568}, {'batch_size': 3, 'throughput': 1.0882073627316602, 'latency_mean': 2.7534224820137023, 'latency_p50': 2.7616323232650757, 'latency_p90': 3.1327243089675902}, {'batch_size': 5, 'throughput': 1.2406255555641532, 'latency_mean': 4.013156788349152, 'latency_p50': 4.021720051765442, 'latency_p90': 4.4970884084701535}, {'batch_size': 6, 'throughput': 1.252900883369407, 'latency_mean': 4.753998689651489, 'latency_p50': 4.7646872997283936, 'latency_p90': 5.334369421005249}, {'batch_size': 8, 'throughput': 1.2486199715572246, 'latency_mean': 6.377750829458237, 'latency_p50': 6.3767207860946655, 'latency_p90': 7.228785181045533}, {'batch_size': 10, 'throughput': 1.2131839144247731, 'latency_mean': 8.186402390003204, 'latency_p50': 8.200952291488647, 'latency_p90': 9.201326751708985}]

max_input_tokens: 1024

max_output_tokens: 64

model_architecture: MistralForCausalLM

model_group: ChaiML/albert_dpo_0912_v

model_name: dpo-ai-selected

model_num_parameters: 12772070400.0

model_repo: ChaiML/albert_dpo_0912_virgo_edit_ai_selected

model_size: 13B

num_battles: 10349

num_wins: 5103

propriety_score: 0.7566079295154186

propriety_total_count: 908.0

ranking_group: single

status: inactive

submission_type: basic

throughput_3p7s: 1.22

timestamp: 2024-09-13T19:37:26+00:00

us_pacific_date: 2024-09-13

win_ratio: 0.4930911199149676

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-v2-mkmlizer
Waiting for job on chaiml-albert-dpo-0912-v-8296-v2-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
Failed to get response for submission mistralai-mixtral-8x7b_3473_v136: ('http://mistralai-mixtral-8x7b-3473-v136-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:47432->127.0.0.1:8080: read: connection reset by peer\n')
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║     _____            __           __                                ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║       /___/                                                         ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║  Version: 0.10.1                                                    ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║  belonging to:                                                      ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: Downloaded to shared memory in 60.705s
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp5ra_tzz9, device:0
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: quantized model in 40.620s
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: Processed model ChaiML/albert_dpo_0912_virgo_edit_ai_selected in 101.325s
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: creating bucket guanaco-mkml-models
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v2
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v2/config.json
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v2/special_tokens_map.json
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v2/tokenizer_config.json
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v2/tokenizer.json
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-8296-v2/flywheel_model.0.safetensors
chaiml-albert-dpo-0912-v-8296-v2-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:15, 22.44it/s]
Loading 0:   3%|▎         | 10/363 [00:00<00:12, 28.86it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:13, 25.03it/s]
Loading 0:   6%|▌         | 21/363 [00:00<00:09, 35.78it/s]
Loading 0:   7%|▋         | 26/363 [00:01<00:16, 20.22it/s]
Loading 0:   9%|▊         | 31/363 [00:01<00:13, 24.84it/s]
Loading 0:  10%|▉         | 35/363 [00:01<00:12, 26.43it/s]
Loading 0:  11%|█         | 39/363 [00:01<00:12, 26.75it/s]
Loading 0:  12%|█▏        | 43/363 [00:01<00:12, 26.53it/s]
Loading 0:  13%|█▎        | 48/363 [00:01<00:10, 29.61it/s]
Loading 0:  14%|█▍        | 52/363 [00:01<00:10, 28.43it/s]
Loading 0:  15%|█▌        | 56/363 [00:02<00:10, 29.37it/s]
Loading 0:  17%|█▋        | 61/363 [00:02<00:12, 25.04it/s]
Loading 0:  18%|█▊        | 64/363 [00:02<00:13, 21.66it/s]
Loading 0:  20%|█▉        | 71/363 [00:02<00:10, 28.76it/s]
Loading 0:  21%|██        | 75/363 [00:02<00:10, 28.51it/s]
Loading 0:  22%|██▏       | 79/363 [00:02<00:10, 27.71it/s]
Loading 0:  23%|██▎       | 84/363 [00:03<00:09, 30.41it/s]
Loading 0:  24%|██▍       | 88/363 [00:03<00:09, 29.66it/s]
Loading 0:  26%|██▌       | 93/363 [00:03<00:08, 32.56it/s]
Loading 0:  27%|██▋       | 97/363 [00:03<00:08, 30.64it/s]
Loading 0:  28%|██▊       | 101/363 [00:03<00:10, 25.16it/s]
Loading 0:  29%|██▊       | 104/363 [00:03<00:11, 22.18it/s]
Loading 0:  31%|███       | 111/363 [00:04<00:08, 29.36it/s]
Loading 0:  32%|███▏      | 115/363 [00:04<00:08, 28.54it/s]
Loading 0:  33%|███▎      | 120/363 [00:04<00:07, 31.24it/s]
Loading 0:  34%|███▍      | 124/363 [00:04<00:07, 30.34it/s]
Loading 0:  36%|███▌      | 129/363 [00:04<00:07, 32.83it/s]
Loading 0:  37%|███▋      | 133/363 [00:04<00:07, 31.00it/s]
Loading 0:  38%|███▊      | 137/363 [00:04<00:07, 31.81it/s]
Loading 0:  39%|███▉      | 142/363 [00:05<00:08, 27.15it/s]
Loading 0:  40%|███▉      | 145/363 [00:05<00:08, 25.29it/s]
Loading 0:  41%|████      | 149/363 [00:05<00:08, 23.90it/s]
Loading 0:  43%|████▎     | 156/363 [00:05<00:06, 30.72it/s]
Loading 0:  44%|████▍     | 160/363 [00:05<00:06, 30.01it/s]
Loading 0:  45%|████▌     | 165/363 [00:05<00:06, 32.58it/s]
Loading 0:  47%|████▋     | 169/363 [00:06<00:06, 30.82it/s]
Loading 0:  48%|████▊     | 174/363 [00:06<00:05, 33.23it/s]
Loading 0:  49%|████▉     | 178/363 [00:06<00:05, 31.78it/s]
Loading 0:  50%|█████     | 182/363 [00:06<00:06, 25.92it/s]
Loading 0:  51%|█████     | 185/363 [00:06<00:07, 22.76it/s]
Loading 0:  53%|█████▎    | 192/363 [00:06<00:05, 30.07it/s]
Loading 0:  54%|█████▍    | 196/363 [00:06<00:05, 29.41it/s]
Loading 0:  55%|█████▌    | 201/363 [00:07<00:05, 31.77it/s]
Loading 0:  56%|█████▋    | 205/363 [00:07<00:05, 30.40it/s]
Loading 0:  58%|█████▊    | 210/363 [00:07<00:04, 32.71it/s]
Loading 0:  59%|█████▉    | 214/363 [00:07<00:04, 30.90it/s]
Loading 0:  60%|██████    | 218/363 [00:07<00:04, 31.29it/s]
Loading 0:  61%|██████▏   | 223/363 [00:07<00:05, 27.10it/s]
Loading 0:  62%|██████▏   | 226/363 [00:08<00:05, 25.34it/s]
Loading 0:  63%|██████▎   | 230/363 [00:08<00:05, 23.76it/s]
Loading 0:  65%|██████▌   | 237/363 [00:08<00:04, 31.04it/s]
Loading 0:  66%|██████▋   | 241/363 [00:08<00:04, 29.78it/s]
Loading 0:  68%|██████▊   | 246/363 [00:08<00:03, 32.25it/s]
Loading 0:  69%|██████▉   | 250/363 [00:08<00:03, 30.30it/s]
Loading 0:  70%|███████   | 255/363 [00:08<00:03, 32.23it/s]
Loading 0:  71%|███████▏  | 259/363 [00:09<00:03, 30.41it/s]
Loading 0:  72%|███████▏  | 263/363 [00:09<00:03, 25.26it/s]
Loading 0:  73%|███████▎  | 266/363 [00:09<00:04, 21.98it/s]
Loading 0:  75%|███████▌  | 273/363 [00:09<00:03, 29.13it/s]
Loading 0:  76%|███████▋  | 277/363 [00:09<00:03, 27.99it/s]
Loading 0:  78%|███████▊  | 282/363 [00:09<00:02, 30.21it/s]
Loading 0:  79%|███████▉  | 286/363 [00:10<00:02, 29.12it/s]
Loading 0:  80%|████████  | 291/363 [00:10<00:02, 31.87it/s]
Loading 0:  81%|████████▏ | 295/363 [00:10<00:02, 30.20it/s]
Loading 0:  82%|████████▏ | 299/363 [00:10<00:02, 30.95it/s]
Loading 0:  84%|████████▎ | 304/363 [00:10<00:02, 26.91it/s]
Loading 0:  85%|████████▍ | 307/363 [00:10<00:02, 25.21it/s]
Loading 0:  86%|████████▌ | 311/363 [00:11<00:02, 23.96it/s]
Loading 0:  88%|████████▊ | 318/363 [00:11<00:01, 31.08it/s]
Loading 0:  89%|████████▊ | 322/363 [00:11<00:01, 28.89it/s]
Loading 0:  90%|█████████ | 327/363 [00:11<00:01, 30.29it/s]
Loading 0:  91%|█████████ | 331/363 [00:11<00:01, 28.78it/s]
Loading 0:  93%|█████████▎| 336/363 [00:11<00:00, 31.00it/s]
Loading 0:  94%|█████████▎| 340/363 [00:11<00:00, 29.53it/s]
Loading 0:  95%|█████████▍| 344/363 [00:18<00:09,  1.98it/s]
Loading 0:  96%|█████████▌| 348/363 [00:19<00:05,  2.66it/s]
Loading 0:  97%|█████████▋| 353/363 [00:19<00:02,  3.86it/s]
Loading 0:  98%|█████████▊| 357/363 [00:19<00:01,  5.02it/s]
                                                            
Job chaiml-albert-dpo-0912-v-8296-v2-mkmlizer completed after 126.65s with status: succeeded
Stopping job with name chaiml-albert-dpo-0912-v-8296-v2-mkmlizer
Pipeline stage MKMLizer completed in 127.54s
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 chaiml-albert-dpo-0912-v-8296-v2
Waiting for inference service chaiml-albert-dpo-0912-v-8296-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
Inference service chaiml-albert-dpo-0912-v-8296-v2 ready after 191.08416056632996s
Pipeline stage MKMLDeployer completed in 191.78s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.1956369876861572s
Received healthy response to inference request in 1.7417781352996826s
Received healthy response to inference request in 1.6291606426239014s
Received healthy response to inference request in 1.621764898300171s
Received healthy response to inference request in 1.6238820552825928s
5 requests
0 failed requests
5th percentile: 1.6221883296966553
10th percentile: 1.6226117610931396
20th percentile: 1.6234586238861084
30th percentile: 1.6249377727508545
40th percentile: 1.627049207687378
50th percentile: 1.6291606426239014
60th percentile: 1.6742076396942138
70th percentile: 1.7192546367645263
80th percentile: 2.032549905776978
90th percentile: 2.6140934467315677
95th percentile: 2.904865217208862
99th percentile: 3.137482633590698
mean time: 1.962444543838501
Pipeline stage StressChecker completed in 10.57s
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 5.42s
Shutdown handler de-registered
chaiml-albert-dpo-0912-v_8296_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 0.11s
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-v2-profiler
Waiting for inference service chaiml-albert-dpo-0912-v-8296-v2-profiler to be ready
Inference service chaiml-albert-dpo-0912-v-8296-v2-profiler ready after 190.5026524066925s
Pipeline stage MKMLProfilerDeployer completed in 190.88s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-albert-dpo-09321b7e1ccf12b6bb835dbf41b1b92443-deploqh6nr:/code/chaiverse_profiler_1726256810 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-albert-dpo-09321b7e1ccf12b6bb835dbf41b1b92443-deploqh6nr --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726256810 && 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 1024 --output_tokens 64 --summary /code/chaiverse_profiler_1726256810/summary.json'
kubectl exec -it chaiml-albert-dpo-09321b7e1ccf12b6bb835dbf41b1b92443-deploqh6nr --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726256810/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1163.24s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-albert-dpo-0912-v-8296-v2-profiler is running
Tearing down inference service chaiml-albert-dpo-0912-v-8296-v2-profiler
Service chaiml-albert-dpo-0912-v-8296-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.16s
Shutdown handler de-registered
chaiml-albert-dpo-0912-v_8296_v2 status is now inactive due to auto deactivation removed underperforming models