submission_id: sao10k-mn-12b-lyra-v3_v2
developer_uid: sao10k
alignment_samples: 10837
alignment_score: -0.21925705625697386
best_of: 8
celo_rating: 1258.74
display_name: Lyra3inst
formatter: {'memory_template': '[INST]system\n{memory}[/INST]\n', 'prompt_template': '[INST]user\n{prompt}[/INST]\n', 'bot_template': '[INST]assistant\n{bot_name}: {message}[/INST]\n', 'user_template': '[INST]user\n{user_name}: {message}[/INST]\n', 'response_template': '[INST]assistant\n{bot_name}:', 'truncate_by_message': False}
generation_params: {'temperature': 0.75, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '\n\n', '\nYou:', '[/INST]', '<|im_end|>', '</s>'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}
gpu_counts: {'NVIDIA RTX A5000': 1}
is_internal_developer: False
language_model: Sao10K/MN-12B-Lyra-v3
latencies: [{'batch_size': 1, 'throughput': 0.6915473217493219, 'latency_mean': 1.4459416031837464, 'latency_p50': 1.4480394124984741, 'latency_p90': 1.6193057298660278}, {'batch_size': 3, 'throughput': 1.3315630521615796, 'latency_mean': 2.2494939255714415, 'latency_p50': 2.246017098426819, 'latency_p90': 2.5112677335739138}, {'batch_size': 5, 'throughput': 1.5645076089351961, 'latency_mean': 3.1834091329574585, 'latency_p50': 3.189576029777527, 'latency_p90': 3.5237700939178467}, {'batch_size': 6, 'throughput': 1.605978711921107, 'latency_mean': 3.712056382894516, 'latency_p50': 3.749447226524353, 'latency_p90': 4.178958320617675}, {'batch_size': 8, 'throughput': 1.5843217186195524, 'latency_mean': 5.025676231384278, 'latency_p50': 5.068355321884155, 'latency_p90': 5.642350482940674}, {'batch_size': 10, 'throughput': 1.538969622422329, 'latency_mean': 6.460110936164856, 'latency_p50': 6.384837985038757, 'latency_p90': 7.4223878383636475}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: MistralForCausalLM
model_group: Sao10K/MN-12B-Lyra-v3
model_name: Lyra3inst
model_num_parameters: 12772070400.0
model_repo: Sao10K/MN-12B-Lyra-v3
model_size: 13B
num_battles: 10837
num_wins: 5751
propriety_score: 0.7234490010515248
propriety_total_count: 951.0
ranking_group: single
status: inactive
submission_type: basic
throughput_3p7s: 1.62
timestamp: 2024-08-28T00:09:08+00:00
us_pacific_date: 2024-08-27
win_ratio: 0.5306819230414321
Download Preference Data
Resubmit model
Running pipeline stage MKMLizer
Starting job with name sao10k-mn-12b-lyra-v3-v2-mkmlizer
Waiting for job on sao10k-mn-12b-lyra-v3-v2-mkmlizer to finish
Stopping job with name sao10k-mn-12b-lyra-v3-v2-mkmlizer
%s, retrying in %s seconds...
Starting job with name sao10k-mn-12b-lyra-v3-v2-mkmlizer
Waiting for job on sao10k-mn-12b-lyra-v3-v2-mkmlizer to finish
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║     _____            __           __                                ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║       /___/                                                         ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║                                                                     ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║  Version: 0.10.1                                                    ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║  https://mk1.ai                                                     ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║                                                                     ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║  belonging to:                                                      ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║                                                                     ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║  Chai Research Corp.                                                ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ║                                                                     ║
sao10k-mn-12b-lyra-v3-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
sao10k-mn-12b-lyra-v3-v2-mkmlizer: Downloaded to shared memory in 50.772s
sao10k-mn-12b-lyra-v3-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp70ov9xcq, device:0
sao10k-mn-12b-lyra-v3-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
sao10k-mn-12b-lyra-v3-v2-mkmlizer: quantized model in 37.020s
sao10k-mn-12b-lyra-v3-v2-mkmlizer: Processed model Sao10K/MN-12B-Lyra-v3 in 87.792s
sao10k-mn-12b-lyra-v3-v2-mkmlizer: creating bucket guanaco-mkml-models
sao10k-mn-12b-lyra-v3-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
sao10k-mn-12b-lyra-v3-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v3-v2
sao10k-mn-12b-lyra-v3-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v3-v2/config.json
sao10k-mn-12b-lyra-v3-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v3-v2/special_tokens_map.json
sao10k-mn-12b-lyra-v3-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v3-v2/tokenizer_config.json
sao10k-mn-12b-lyra-v3-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v3-v2/tokenizer.json
sao10k-mn-12b-lyra-v3-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v3-v2/flywheel_model.0.safetensors
sao10k-mn-12b-lyra-v3-v2-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:10, 32.83it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:07, 47.17it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:10, 33.16it/s]
Loading 0:   6%|▋         | 23/363 [00:00<00:10, 32.25it/s]
Loading 0:   8%|▊         | 30/363 [00:00<00:08, 40.20it/s]
Loading 0:  10%|▉         | 35/363 [00:00<00:08, 40.81it/s]
Loading 0:  11%|█         | 40/363 [00:01<00:08, 37.39it/s]
Loading 0:  12%|█▏        | 44/363 [00:01<00:10, 31.14it/s]
Loading 0:  14%|█▍        | 50/363 [00:01<00:09, 33.14it/s]
Loading 0:  16%|█▌        | 57/363 [00:01<00:07, 39.87it/s]
Loading 0:  17%|█▋        | 62/363 [00:01<00:07, 40.69it/s]
Loading 0:  18%|█▊        | 67/363 [00:01<00:06, 42.65it/s]
Loading 0:  20%|█▉        | 72/363 [00:01<00:06, 42.39it/s]
Loading 0:  21%|██        | 77/363 [00:02<00:07, 37.97it/s]
Loading 0:  22%|██▏       | 81/363 [00:02<00:09, 30.39it/s]
Loading 0:  24%|██▍       | 88/363 [00:02<00:08, 33.90it/s]
Loading 0:  26%|██▌       | 94/363 [00:02<00:07, 38.18it/s]
Loading 0:  27%|██▋       | 99/363 [00:02<00:07, 36.81it/s]
Loading 0:  28%|██▊       | 103/363 [00:02<00:07, 36.62it/s]
Loading 0:  29%|██▉       | 107/363 [00:02<00:07, 33.01it/s]
Loading 0:  31%|███       | 113/363 [00:03<00:07, 34.42it/s]
Loading 0:  33%|███▎      | 120/363 [00:03<00:06, 40.11it/s]
Loading 0:  34%|███▍      | 125/363 [00:03<00:05, 39.75it/s]
Loading 0:  36%|███▌      | 130/363 [00:03<00:05, 41.36it/s]
Loading 0:  37%|███▋      | 136/363 [00:03<00:05, 41.46it/s]
Loading 0:  39%|███▉      | 141/363 [00:03<00:05, 37.28it/s]
Loading 0:  40%|███▉      | 145/363 [00:03<00:06, 34.52it/s]
Loading 0:  41%|████      | 149/363 [00:04<00:06, 32.97it/s]
Loading 0:  43%|████▎     | 156/363 [00:04<00:04, 41.43it/s]
Loading 0:  44%|████▍     | 161/363 [00:04<00:04, 40.50it/s]
Loading 0:  46%|████▌     | 166/363 [00:04<00:05, 34.08it/s]
Loading 0:  47%|████▋     | 170/363 [00:04<00:06, 31.76it/s]
Loading 0:  48%|████▊     | 176/363 [00:04<00:05, 33.79it/s]
Loading 0:  50%|█████     | 183/363 [00:04<00:04, 40.94it/s]
Loading 0:  52%|█████▏    | 188/363 [00:05<00:04, 41.52it/s]
Loading 0:  53%|█████▎    | 193/363 [00:05<00:04, 40.19it/s]
Loading 0:  55%|█████▍    | 198/363 [00:05<00:04, 41.20it/s]
Loading 0:  56%|█████▌    | 203/363 [00:05<00:04, 39.47it/s]
Loading 0:  57%|█████▋    | 208/363 [00:05<00:04, 31.07it/s]
Loading 0:  58%|█████▊    | 212/363 [00:05<00:04, 30.62it/s]
Loading 0:  61%|██████    | 220/363 [00:05<00:03, 39.51it/s]
Loading 0:  62%|██████▏   | 225/363 [00:06<00:03, 41.03it/s]
Loading 0:  63%|██████▎   | 230/363 [00:06<00:03, 37.65it/s]
Loading 0:  64%|██████▍   | 234/363 [00:06<00:04, 29.93it/s]
Loading 0:  66%|██████▌   | 239/363 [00:06<00:03, 32.18it/s]
Loading 0:  68%|██████▊   | 247/363 [00:06<00:02, 40.77it/s]
Loading 0:  69%|██████▉   | 252/363 [00:06<00:02, 41.31it/s]
Loading 0:  71%|███████   | 257/363 [00:06<00:02, 35.84it/s]
Loading 0:  73%|███████▎  | 266/363 [00:07<00:02, 42.19it/s]
Loading 0:  75%|███████▍  | 271/363 [00:07<00:02, 33.33it/s]
Loading 0:  76%|███████▌  | 275/363 [00:07<00:02, 31.83it/s]
Loading 0:  78%|███████▊  | 282/363 [00:07<00:02, 38.93it/s]
Loading 0:  79%|███████▉  | 287/363 [00:07<00:01, 40.67it/s]
Loading 0:  80%|████████  | 292/363 [00:07<00:01, 35.82it/s]
Loading 0:  82%|████████▏ | 296/363 [00:08<00:02, 30.88it/s]
Loading 0:  83%|████████▎ | 302/363 [00:08<00:01, 33.82it/s]
Loading 0:  85%|████████▌ | 309/363 [00:08<00:01, 41.57it/s]
Loading 0:  87%|████████▋ | 314/363 [00:08<00:01, 42.33it/s]
Loading 0:  88%|████████▊ | 319/363 [00:08<00:01, 43.25it/s]
Loading 0:  89%|████████▉ | 324/363 [00:08<00:00, 43.52it/s]
Loading 0:  91%|█████████ | 329/363 [00:08<00:00, 41.63it/s]
Loading 0:  92%|█████████▏| 334/363 [00:09<00:00, 33.84it/s]
Loading 0:  93%|█████████▎| 338/363 [00:09<00:00, 31.74it/s]
Loading 0:  95%|█████████▌| 345/363 [00:09<00:00, 39.20it/s]
Loading 0:  96%|█████████▋| 350/363 [00:09<00:00, 39.23it/s]
Loading 0:  98%|█████████▊| 355/363 [00:09<00:00, 34.61it/s]
Loading 0:  99%|█████████▉| 359/363 [00:16<00:01,  2.30it/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
Job sao10k-mn-12b-lyra-v3-v2-mkmlizer completed after 115.42s with status: succeeded
Stopping job with name sao10k-mn-12b-lyra-v3-v2-mkmlizer
Pipeline stage MKMLizer completed in 117.39s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.43s
Running pipeline stage ISVCDeployer
Creating inference service sao10k-mn-12b-lyra-v3-v2
Waiting for inference service sao10k-mn-12b-lyra-v3-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
Inference service sao10k-mn-12b-lyra-v3-v2 ready after 170.73763847351074s
Pipeline stage ISVCDeployer completed in 171.30s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.6374621391296387s
Received healthy response to inference request in 2.022188663482666s
Received healthy response to inference request in 2.1008050441741943s
Received healthy response to inference request in 2.1720197200775146s
Received healthy response to inference request in 2.750924825668335s
5 requests
0 failed requests
5th percentile: 2.0379119396209715
10th percentile: 2.0536352157592774
20th percentile: 2.085081768035889
30th percentile: 2.115047979354858
40th percentile: 2.1435338497161864
50th percentile: 2.1720197200775146
60th percentile: 2.3581966876983644
70th percentile: 2.544373655319214
80th percentile: 2.660154676437378
90th percentile: 2.7055397510528563
95th percentile: 2.7282322883605956
99th percentile: 2.746386318206787
mean time: 2.3366800785064696
Pipeline stage StressChecker completed in 12.68s
sao10k-mn-12b-lyra-v3_v2 status is now deployed due to DeploymentManager action
sao10k-mn-12b-lyra-v3_v2 status is now inactive due to auto deactivation removed underperforming models
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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service sao10k-mn-12b-lyra-v3-v2-profiler
Waiting for inference service sao10k-mn-12b-lyra-v3-v2-profiler to be ready
Inference service sao10k-mn-12b-lyra-v3-v2-profiler ready after 140.3196358680725s
Pipeline stage MKMLProfilerDeployer completed in 140.71s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/sao10k-mn-12b-lyra-v3-v2-profiler-predictor-00001-deploymejxqpf:/code/chaiverse_profiler_1725502255 --namespace tenant-chaiml-guanaco
kubectl exec -it sao10k-mn-12b-lyra-v3-v2-profiler-predictor-00001-deploymejxqpf --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725502255 && 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_1725502255/summary.json'
kubectl exec -it sao10k-mn-12b-lyra-v3-v2-profiler-predictor-00001-deploymejxqpf --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725502255/summary.json'
Pipeline stage MKMLProfilerRunner completed in 946.55s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service sao10k-mn-12b-lyra-v3-v2-profiler is running
Tearing down inference service sao10k-mn-12b-lyra-v3-v2-profiler
Service sao10k-mn-12b-lyra-v3-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.84s
Shutdown handler de-registered
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service sao10k-mn-12b-lyra-v3-v2-profiler is running
Skipping teardown as no inference service was found
Pipeline stage MKMLProfilerDeleter completed in 1.84s
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 sao10k-mn-12b-lyra-v3-v2-profiler
Waiting for inference service sao10k-mn-12b-lyra-v3-v2-profiler to be ready
Inference service sao10k-mn-12b-lyra-v3-v2-profiler ready after 150.3677372932434s
Pipeline stage MKMLProfilerDeployer completed in 150.73s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/sao10k-mn-12b-lyra-v3-v2-profiler-predictor-00001-deployme2tv5c:/code/chaiverse_profiler_1725503723 --namespace tenant-chaiml-guanaco
kubectl exec -it sao10k-mn-12b-lyra-v3-v2-profiler-predictor-00001-deployme2tv5c --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725503723 && 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_1725503723/summary.json'
kubectl exec -it sao10k-mn-12b-lyra-v3-v2-profiler-predictor-00001-deployme2tv5c --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725503723/summary.json'
Pipeline stage MKMLProfilerRunner completed in 953.08s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service sao10k-mn-12b-lyra-v3-v2-profiler is running
Tearing down inference service sao10k-mn-12b-lyra-v3-v2-profiler
Service sao10k-mn-12b-lyra-v3-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.68s
Shutdown handler de-registered

Usage Metrics

Latency Metrics