rica40325-mistral-7356

submission_id: rica40325-mistral-7356_v1

developer_uid: rica40325

best_of: 8

celo_rating: 1260.85

display_name: rica40325-mistral-7356_v1

family_friendly_score: 0.5571536714610144

family_friendly_standard_error: 0.009643355384882526

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}

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

ineligible_reason: num_battles<5000

is_internal_developer: False

language_model: rica40325/mistral-7356

latencies: [{'batch_size': 1, 'throughput': 0.6234470526646358, 'latency_mean': 1.6039264273643494, 'latency_p50': 1.6004087924957275, 'latency_p90': 1.776209020614624}, {'batch_size': 3, 'throughput': 1.0977026732550714, 'latency_mean': 2.731638487577438, 'latency_p50': 2.728041648864746, 'latency_p90': 3.054158639907837}, {'batch_size': 5, 'throughput': 1.2528262191871071, 'latency_mean': 3.978855959177017, 'latency_p50': 3.989555597305298, 'latency_p90': 4.451220703125}, {'batch_size': 6, 'throughput': 1.2699708020689715, 'latency_mean': 4.7096017599105835, 'latency_p50': 4.725161790847778, 'latency_p90': 5.298621129989624}, {'batch_size': 8, 'throughput': 1.2569204314865907, 'latency_mean': 6.322893973588943, 'latency_p50': 6.400106906890869, 'latency_p90': 7.055029439926147}, {'batch_size': 10, 'throughput': 1.2158661742951093, 'latency_mean': 8.181758905649184, 'latency_p50': 8.250110507011414, 'latency_p90': 9.334281921386719}]

max_input_tokens: 1024

max_output_tokens: 64

model_architecture: MistralForCausalLM

model_group: rica40325/mistral-7356

model_name: rica40325-mistral-7356_v1

model_num_parameters: 12772070400.0

model_repo: rica40325/mistral-7356

model_size: 13B

num_battles: 2703

num_wins: 1374

ranking_group: single

status: torndown

submission_type: basic

throughput_3p7s: 1.23

timestamp: 2024-09-25T05:03:51+00:00

us_pacific_date: 2024-09-24

win_ratio: 0.5083240843507214

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 rica40325-mistral-7356-v1-mkmlizer
Waiting for job on rica40325-mistral-7356-v1-mkmlizer to finish
rica40325-mistral-7356-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-mistral-7356-v1-mkmlizer: ║     _____            __           __                                ║
rica40325-mistral-7356-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rica40325-mistral-7356-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rica40325-mistral-7356-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rica40325-mistral-7356-v1-mkmlizer: ║       /___/                                                         ║
rica40325-mistral-7356-v1-mkmlizer: ║                                                                     ║
rica40325-mistral-7356-v1-mkmlizer: ║  Version: 0.11.12                                                   ║
rica40325-mistral-7356-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rica40325-mistral-7356-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rica40325-mistral-7356-v1-mkmlizer: ║                                                                     ║
rica40325-mistral-7356-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rica40325-mistral-7356-v1-mkmlizer: ║  belonging to:                                                      ║
rica40325-mistral-7356-v1-mkmlizer: ║                                                                     ║
rica40325-mistral-7356-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rica40325-mistral-7356-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rica40325-mistral-7356-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rica40325-mistral-7356-v1-mkmlizer: ║                                                                     ║
rica40325-mistral-7356-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
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
rica40325-mistral-7356-v1-mkmlizer: Downloaded to shared memory in 52.028s
rica40325-mistral-7356-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpxumnfnnb, device:0
rica40325-mistral-7356-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-mistral-7356-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/mk1/flywheel/functional/loader.py:55: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
rica40325-mistral-7356-v1-mkmlizer:   tensors = torch.load(model_shard_filename, map_location=torch.device(self.device), mmap=True)
rica40325-mistral-7356-v1-mkmlizer: quantized model in 36.611s
rica40325-mistral-7356-v1-mkmlizer: Processed model rica40325/mistral-7356 in 88.639s
rica40325-mistral-7356-v1-mkmlizer: creating bucket guanaco-mkml-models
rica40325-mistral-7356-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-mistral-7356-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-mistral-7356-v1
rica40325-mistral-7356-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-mistral-7356-v1/config.json
rica40325-mistral-7356-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-mistral-7356-v1/special_tokens_map.json
rica40325-mistral-7356-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-mistral-7356-v1/tokenizer_config.json
rica40325-mistral-7356-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-mistral-7356-v1/tokenizer.json
rica40325-mistral-7356-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-mistral-7356-v1/flywheel_model.0.safetensors
rica40325-mistral-7356-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/363 [00:00<00:07, 49.46it/s]
Loading 0:   6%|▌         | 22/363 [00:00<00:04, 80.35it/s]
Loading 0:   9%|▉         | 34/363 [00:00<00:04, 81.73it/s]
Loading 0:  12%|█▏        | 44/363 [00:00<00:03, 86.11it/s]
Loading 0:  15%|█▍        | 53/363 [00:00<00:03, 85.03it/s]
Loading 0:  17%|█▋        | 62/363 [00:01<00:14, 20.10it/s]
Loading 0:  19%|█▉        | 70/363 [00:01<00:11, 25.36it/s]
Loading 0:  22%|██▏       | 80/363 [00:02<00:08, 33.75it/s]
Loading 0:  24%|██▍       | 88/363 [00:02<00:06, 39.62it/s]
Loading 0:  27%|██▋       | 97/363 [00:02<00:05, 47.59it/s]
Loading 0:  29%|██▉       | 106/363 [00:02<00:04, 52.56it/s]
Loading 0:  32%|███▏      | 115/363 [00:02<00:04, 58.45it/s]
Loading 0:  34%|███▍      | 125/363 [00:02<00:03, 67.22it/s]
Loading 0:  37%|███▋      | 134/363 [00:02<00:03, 70.86it/s]
Loading 0:  39%|███▉      | 143/363 [00:03<00:10, 20.00it/s]
Loading 0:  42%|████▏     | 152/363 [00:04<00:08, 25.97it/s]
Loading 0:  44%|████▍     | 160/363 [00:04<00:06, 31.53it/s]
Loading 0:  47%|████▋     | 169/363 [00:04<00:05, 38.41it/s]
Loading 0:  49%|████▉     | 178/363 [00:04<00:03, 46.38it/s]
Loading 0:  52%|█████▏    | 187/363 [00:04<00:03, 54.26it/s]
Loading 0:  54%|█████▍    | 196/363 [00:04<00:02, 61.38it/s]
Loading 0:  56%|█████▋    | 205/363 [00:04<00:02, 66.42it/s]
Loading 0:  59%|█████▉    | 215/363 [00:04<00:01, 74.02it/s]
Loading 0:  62%|██████▏   | 224/363 [00:05<00:06, 21.24it/s]
Loading 0:  64%|██████▍   | 232/363 [00:06<00:04, 26.42it/s]
Loading 0:  66%|██████▋   | 241/363 [00:06<00:03, 32.76it/s]
Loading 0:  69%|██████▉   | 250/363 [00:06<00:02, 40.55it/s]
Loading 0:  71%|███████▏  | 259/363 [00:06<00:02, 47.41it/s]
Loading 0:  74%|███████▍  | 268/363 [00:06<00:01, 54.02it/s]
Loading 0:  76%|███████▋  | 277/363 [00:06<00:01, 60.57it/s]
Loading 0:  79%|███████▉  | 286/363 [00:06<00:01, 65.88it/s]
Loading 0:  81%|████████▏ | 295/363 [00:06<00:00, 70.91it/s]
Loading 0:  84%|████████▎ | 304/363 [00:07<00:02, 20.95it/s]
Loading 0:  86%|████████▌ | 313/363 [00:08<00:01, 27.21it/s]
Loading 0:  89%|████████▉ | 323/363 [00:08<00:01, 35.45it/s]
Loading 0:  91%|█████████ | 331/363 [00:08<00:00, 41.35it/s]
Loading 0:  94%|█████████▎| 340/363 [00:08<00:00, 49.00it/s]
Loading 0:  96%|█████████▋| 350/363 [00:08<00:00, 58.43it/s]
Loading 0:  99%|█████████▉| 359/363 [00:08<00:00, 65.07it/s]
                                                            
Job rica40325-mistral-7356-v1-mkmlizer completed after 113.83s with status: succeeded
Stopping job with name rica40325-mistral-7356-v1-mkmlizer
Pipeline stage MKMLizer completed in 114.91s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.10s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rica40325-mistral-7356-v1
Waiting for inference service rica40325-mistral-7356-v1 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
Inference service rica40325-mistral-7356-v1 ready after 211.84163069725037s
Pipeline stage MKMLDeployer completed in 212.31s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.1414988040924072s
Received healthy response to inference request in 2.836707353591919s
Received healthy response to inference request in 2.098956346511841s
Received healthy response to inference request in 4.144035577774048s
Received healthy response to inference request in 2.3403818607330322s
5 requests
0 failed requests
5th percentile: 2.147241449356079
10th percentile: 2.1955265522003176
20th percentile: 2.292096757888794
30th percentile: 2.4396469593048096
40th percentile: 2.6381771564483643
50th percentile: 2.836707353591919
60th percentile: 2.958623933792114
70th percentile: 3.0805405139923097
80th percentile: 3.3420061588287355
90th percentile: 3.7430208683013917
95th percentile: 3.9435282230377195
99th percentile: 4.103934106826782
mean time: 2.9123159885406493
Pipeline stage StressChecker completed in 15.66s
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 5.37s
Shutdown handler de-registered
rica40325-mistral-7356_v1 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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-mistral-7356-v1-profiler
Waiting for inference service rica40325-mistral-7356-v1-profiler to be ready
Inference service rica40325-mistral-7356-v1-profiler ready after 220.51224279403687s
Pipeline stage MKMLProfilerDeployer completed in 220.91s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-mistral-7356-v1-profiler-predictor-00001-deploym82q2k:/code/chaiverse_profiler_1727241249 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-mistral-7356-v1-profiler-predictor-00001-deploym82q2k --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1727241249 && 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_1727241249/summary.json'
kubectl exec -it rica40325-mistral-7356-v1-profiler-predictor-00001-deploym82q2k --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1727241249/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1150.06s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-mistral-7356-v1-profiler is running
Tearing down inference service rica40325-mistral-7356-v1-profiler
Service rica40325-mistral-7356-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.32s
Shutdown handler de-registered
rica40325-mistral-7356_v1 status is now inactive due to auto deactivation removed underperforming models
rica40325-mistral-7356_v1 status is now torndown due to DeploymentManager action