rica40325-mistral-7356

submission_id: rica40325-mistral-7356_v2

developer_uid: rica40325

best_of: 8

celo_rating: 1260.58

display_name: rica40325-mistral-7356_v1

family_friendly_score: 0.5542251111871366

family_friendly_standard_error: 0.009163658998844523

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': ['</s>', 'Bot:', 'User:', 'You:', 'Me:', '####'], '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': 0.6206564320035739, 'latency_mean': 1.6111376106739044, 'latency_p50': 1.6188888549804688, 'latency_p90': 1.767330026626587}, {'batch_size': 3, 'throughput': 1.0943562672869553, 'latency_mean': 2.73918696641922, 'latency_p50': 2.7634856700897217, 'latency_p90': 3.0023796796798705}, {'batch_size': 5, 'throughput': 1.2433726504487328, 'latency_mean': 4.003507262468338, 'latency_p50': 4.026834726333618, 'latency_p90': 4.522751712799073}, {'batch_size': 6, 'throughput': 1.273986983995119, 'latency_mean': 4.685184941291809, 'latency_p50': 4.677178144454956, 'latency_p90': 5.256221747398376}, {'batch_size': 8, 'throughput': 1.2598554503856088, 'latency_mean': 6.309645557403565, 'latency_p50': 6.3839733600616455, 'latency_p90': 7.164608383178711}, {'batch_size': 10, 'throughput': 1.218242367230381, 'latency_mean': 8.164681758880615, 'latency_p50': 8.169075727462769, 'latency_p90': 9.343198704719544}]

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: 3003

num_wins: 1538

ranking_group: single

status: torndown

submission_type: basic

throughput_3p7s: 1.22

timestamp: 2024-09-25T06:00:45+00:00

us_pacific_date: 2024-09-24

win_ratio: 0.5121545121545121

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-v2-mkmlizer
Waiting for job on rica40325-mistral-7356-v2-mkmlizer to finish
rica40325-mistral-7356-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-mistral-7356-v2-mkmlizer: ║     _____            __           __                                ║
rica40325-mistral-7356-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rica40325-mistral-7356-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rica40325-mistral-7356-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rica40325-mistral-7356-v2-mkmlizer: ║       /___/                                                         ║
rica40325-mistral-7356-v2-mkmlizer: ║                                                                     ║
rica40325-mistral-7356-v2-mkmlizer: ║  Version: 0.11.12                                                   ║
rica40325-mistral-7356-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
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-v2-mkmlizer: ║  https://mk1.ai                                                     ║
rica40325-mistral-7356-v2-mkmlizer: ║                                                                     ║
rica40325-mistral-7356-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
rica40325-mistral-7356-v2-mkmlizer: ║  belonging to:                                                      ║
rica40325-mistral-7356-v2-mkmlizer: ║                                                                     ║
rica40325-mistral-7356-v2-mkmlizer: ║  Chai Research Corp.                                                ║
rica40325-mistral-7356-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rica40325-mistral-7356-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rica40325-mistral-7356-v2-mkmlizer: ║                                                                     ║
rica40325-mistral-7356-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rica40325-mistral-7356-v2-mkmlizer: Downloaded to shared memory in 30.750s
rica40325-mistral-7356-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpr4p9h23u, device:0
rica40325-mistral-7356-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-mistral-7356-v2-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-v2-mkmlizer:   tensors = torch.load(model_shard_filename, map_location=torch.device(self.device), mmap=True)
Connection pool is full, discarding connection: %s. Connection pool size: %s
rica40325-mistral-7356-v2-mkmlizer: quantized model in 35.981s
rica40325-mistral-7356-v2-mkmlizer: Processed model rica40325/mistral-7356 in 66.731s
rica40325-mistral-7356-v2-mkmlizer: creating bucket guanaco-mkml-models
rica40325-mistral-7356-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-mistral-7356-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-mistral-7356-v2
rica40325-mistral-7356-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-mistral-7356-v2/config.json
rica40325-mistral-7356-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-mistral-7356-v2/special_tokens_map.json
rica40325-mistral-7356-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-mistral-7356-v2/tokenizer_config.json
rica40325-mistral-7356-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-mistral-7356-v2/tokenizer.json
rica40325-mistral-7356-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-mistral-7356-v2/flywheel_model.0.safetensors
rica40325-mistral-7356-v2-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/363 [00:00<00:06, 51.29it/s]
Loading 0:   6%|▌         | 22/363 [00:00<00:04, 82.69it/s]
Loading 0:   9%|▉         | 33/363 [00:00<00:03, 92.30it/s]
Loading 0:  12%|█▏        | 43/363 [00:00<00:03, 81.07it/s]
Loading 0:  14%|█▍        | 52/363 [00:00<00:03, 80.58it/s]
Loading 0:  17%|█▋        | 61/363 [00:01<00:14, 20.99it/s]
Loading 0:  20%|█▉        | 71/363 [00:01<00:10, 28.37it/s]
Loading 0:  23%|██▎       | 85/363 [00:02<00:07, 39.55it/s]
Loading 0:  26%|██▌       | 94/363 [00:02<00:05, 46.37it/s]
Loading 0:  28%|██▊       | 103/363 [00:02<00:04, 52.66it/s]
Loading 0:  31%|███       | 113/363 [00:02<00:04, 61.39it/s]
Loading 0:  34%|███▎      | 122/363 [00:02<00:03, 67.09it/s]
Loading 0:  36%|███▋      | 132/363 [00:02<00:03, 74.69it/s]
Loading 0:  39%|███▉      | 142/363 [00:03<00:10, 20.62it/s]
Loading 0:  42%|████▏     | 152/363 [00:03<00:07, 27.12it/s]
Loading 0:  44%|████▍     | 160/363 [00:04<00:06, 31.72it/s]
Loading 0:  47%|████▋     | 169/363 [00:04<00:05, 37.86it/s]
Loading 0:  49%|████▉     | 178/363 [00:04<00:04, 44.05it/s]
Loading 0:  52%|█████▏    | 187/363 [00:04<00:03, 51.22it/s]
Loading 0:  54%|█████▍    | 196/363 [00:04<00:02, 56.67it/s]
Loading 0:  56%|█████▋    | 205/363 [00:04<00:02, 62.61it/s]
Loading 0:  59%|█████▉    | 214/363 [00:04<00:02, 68.37it/s]
Loading 0:  61%|██████▏   | 223/363 [00:05<00:06, 21.41it/s]
Loading 0:  64%|██████▍   | 232/363 [00:05<00:04, 27.24it/s]
Loading 0:  67%|██████▋   | 242/363 [00:06<00:03, 35.47it/s]
Loading 0:  69%|██████▉   | 251/363 [00:06<00:02, 43.05it/s]
Loading 0:  71%|███████▏  | 259/363 [00:06<00:02, 49.08it/s]
Loading 0:  74%|███████▍  | 268/363 [00:06<00:01, 55.08it/s]
Loading 0:  76%|███████▋  | 277/363 [00:06<00:01, 61.67it/s]
Loading 0:  79%|███████▉  | 286/363 [00:06<00:01, 66.75it/s]
Loading 0:  81%|████████▏ | 295/363 [00:06<00:00, 71.35it/s]
Loading 0:  84%|████████▎ | 304/363 [00:07<00:02, 21.05it/s]
Loading 0:  86%|████████▌ | 313/363 [00:07<00:01, 26.49it/s]
Loading 0:  89%|████████▊ | 322/363 [00:08<00:01, 33.59it/s]
Loading 0:  91%|█████████ | 331/363 [00:08<00:00, 39.90it/s]
Loading 0:  94%|█████████▎| 340/363 [00:08<00:00, 46.84it/s]
Loading 0:  96%|█████████▌| 349/363 [00:08<00:00, 53.93it/s]
Loading 0:  99%|█████████▊| 358/363 [00:08<00:00, 60.82it/s]
                                                            
Job rica40325-mistral-7356-v2-mkmlizer completed after 88.27s with status: succeeded
Stopping job with name rica40325-mistral-7356-v2-mkmlizer
Pipeline stage MKMLizer completed in 89.18s
run pipeline stage %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
Running pipeline stage MKMLTemplater
Connection pool is full, discarding connection: %s. Connection pool size: %s
Pipeline stage MKMLTemplater completed in 0.22s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rica40325-mistral-7356-v2
Waiting for inference service rica40325-mistral-7356-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 rica40325-mistral-7356-v2 ready after 212.89391207695007s
Pipeline stage MKMLDeployer completed in 213.28s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 4.107136249542236s
Received healthy response to inference request in 2.2780020236968994s
Received healthy response to inference request in 2.17486310005188s
Received healthy response to inference request in 2.40856671333313s
Received healthy response to inference request in 3.75579833984375s
5 requests
0 failed requests
5th percentile: 2.1954908847808836
10th percentile: 2.216118669509888
20th percentile: 2.2573742389678957
30th percentile: 2.3041149616241454
40th percentile: 2.356340837478638
50th percentile: 2.40856671333313
60th percentile: 2.947459363937378
70th percentile: 3.486352014541626
80th percentile: 3.826065921783447
90th percentile: 3.966601085662842
95th percentile: 4.036868667602539
99th percentile: 4.093082733154297
mean time: 2.9448732852935793
Pipeline stage StressChecker completed in 17.25s
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 5.63s
Shutdown handler de-registered
rica40325-mistral-7356_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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-mistral-7356-v2-profiler
Waiting for inference service rica40325-mistral-7356-v2-profiler to be ready
Inference service rica40325-mistral-7356-v2-profiler ready after 210.6205518245697s
Pipeline stage MKMLProfilerDeployer completed in 210.99s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-mistral-7356-v2-profiler-predictor-00001-deploymlvpn2:/code/chaiverse_profiler_1727244627 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-mistral-7356-v2-profiler-predictor-00001-deploymlvpn2 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1727244627 && 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_1727244627/summary.json'
kubectl exec -it rica40325-mistral-7356-v2-profiler-predictor-00001-deploymlvpn2 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1727244627/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1152.37s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-mistral-7356-v2-profiler is running
Tearing down inference service rica40325-mistral-7356-v2-profiler
Service rica40325-mistral-7356-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.23s
Shutdown handler de-registered
rica40325-mistral-7356_v2 status is now inactive due to auto deactivation removed underperforming models
rica40325-mistral-7356_v2 status is now torndown due to DeploymentManager action