rinen0721-llama0914-1426

submission_id: rinen0721-llama0914-1426_v3

developer_uid: rinen0721

best_of: 16

celo_rating: 1221.13

display_name: rinen0721-llama0914-1426_v3

family_friendly_score: 0.0

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': 512, 'best_of': 16, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

is_internal_developer: False

language_model: rinen0721/llama0914-1426

latencies: [{'batch_size': 1, 'throughput': 0.9067354730862742, 'latency_mean': 1.1027597284317017, 'latency_p50': 1.096766710281372, 'latency_p90': 1.2379829645156861}, {'batch_size': 4, 'throughput': 1.7969264558306244, 'latency_mean': 2.218116030693054, 'latency_p50': 2.1952418088912964, 'latency_p90': 2.4944038867950438}, {'batch_size': 5, 'throughput': 1.8908695611815862, 'latency_mean': 2.631982307434082, 'latency_p50': 2.6115247011184692, 'latency_p90': 2.9206918478012085}, {'batch_size': 8, 'throughput': 1.980631175785444, 'latency_mean': 4.000624951124191, 'latency_p50': 3.9875353574752808, 'latency_p90': 4.476140832901001}, {'batch_size': 10, 'throughput': 2.0339741788574797, 'latency_mean': 4.875884300470352, 'latency_p50': 4.816362619400024, 'latency_p90': 5.66960506439209}, {'batch_size': 12, 'throughput': 2.0217459798598276, 'latency_mean': 5.856473499536515, 'latency_p50': 5.855260491371155, 'latency_p90': 6.651525688171387}, {'batch_size': 15, 'throughput': 2.0261803771412126, 'latency_mean': 7.264227142333985, 'latency_p50': 7.3898491859436035, 'latency_p90': 8.09384126663208}]

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: rinen0721/llama0914-1426

model_name: rinen0721-llama0914-1426_v3

model_num_parameters: 8030261248.0

model_repo: rinen0721/llama0914-1426

model_size: 8B

num_battles: 10906

num_wins: 4963

ranking_group: single

status: torndown

submission_type: basic

throughput_3p7s: 1.98

timestamp: 2024-09-14T12:52:32+00:00

us_pacific_date: 2024-09-14

win_ratio: 0.45507060333761234

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 rinen0721-llama0914-1426-v3-mkmlizer
Waiting for job on rinen0721-llama0914-1426-v3-mkmlizer to finish
rinen0721-llama0914-1426-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rinen0721-llama0914-1426-v3-mkmlizer: ║     _____            __           __                                ║
rinen0721-llama0914-1426-v3-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rinen0721-llama0914-1426-v3-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rinen0721-llama0914-1426-v3-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rinen0721-llama0914-1426-v3-mkmlizer: ║       /___/                                                         ║
rinen0721-llama0914-1426-v3-mkmlizer: ║                                                                     ║
rinen0721-llama0914-1426-v3-mkmlizer: ║  Version: 0.10.1                                                    ║
rinen0721-llama0914-1426-v3-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rinen0721-llama0914-1426-v3-mkmlizer: ║  https://mk1.ai                                                     ║
rinen0721-llama0914-1426-v3-mkmlizer: ║                                                                     ║
rinen0721-llama0914-1426-v3-mkmlizer: ║  The license key for the current software has been verified as      ║
rinen0721-llama0914-1426-v3-mkmlizer: ║  belonging to:                                                      ║
rinen0721-llama0914-1426-v3-mkmlizer: ║                                                                     ║
rinen0721-llama0914-1426-v3-mkmlizer: ║  Chai Research Corp.                                                ║
rinen0721-llama0914-1426-v3-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rinen0721-llama0914-1426-v3-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rinen0721-llama0914-1426-v3-mkmlizer: ║                                                                     ║
rinen0721-llama0914-1426-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Connection pool is full, discarding connection: %s. Connection pool size: %s
rinen0721-llama0914-1426-v3-mkmlizer: Downloaded to shared memory in 25.095s
rinen0721-llama0914-1426-v3-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpv1y5gno6, device:0
rinen0721-llama0914-1426-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Failed to get response for submission blend_hokok_2024-09-09: ('http://neversleep-noromaid-v0-8068-v150-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '')
rinen0721-llama0914-1426-v3-mkmlizer: quantized model in 26.629s
rinen0721-llama0914-1426-v3-mkmlizer: Processed model rinen0721/llama0914-1426 in 51.724s
rinen0721-llama0914-1426-v3-mkmlizer: creating bucket guanaco-mkml-models
rinen0721-llama0914-1426-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rinen0721-llama0914-1426-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rinen0721-llama0914-1426-v3
rinen0721-llama0914-1426-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rinen0721-llama0914-1426-v3/special_tokens_map.json
rinen0721-llama0914-1426-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rinen0721-llama0914-1426-v3/config.json
rinen0721-llama0914-1426-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rinen0721-llama0914-1426-v3/tokenizer.json
rinen0721-llama0914-1426-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rinen0721-llama0914-1426-v3/flywheel_model.0.safetensors
rinen0721-llama0914-1426-v3-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 54.12it/s]
Loading 0:   8%|▊         | 22/291 [00:00<00:03, 81.10it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:03, 75.63it/s]
Loading 0:  14%|█▎        | 40/291 [00:00<00:03, 80.23it/s]
Loading 0:  17%|█▋        | 49/291 [00:00<00:02, 81.74it/s]
Loading 0:  20%|█▉        | 58/291 [00:00<00:02, 81.70it/s]
Loading 0:  23%|██▎       | 67/291 [00:00<00:02, 83.76it/s]
Loading 0:  26%|██▌       | 76/291 [00:00<00:02, 85.07it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:09, 20.75it/s]
Loading 0:  32%|███▏      | 94/291 [00:02<00:07, 27.01it/s]
Loading 0:  35%|███▌      | 103/291 [00:02<00:05, 33.61it/s]
Loading 0:  38%|███▊      | 112/291 [00:02<00:04, 41.34it/s]
Loading 0:  42%|████▏     | 121/291 [00:02<00:03, 49.11it/s]
Loading 0:  45%|████▍     | 130/291 [00:02<00:02, 55.95it/s]
Loading 0:  48%|████▊     | 139/291 [00:02<00:02, 60.40it/s]
Loading 0:  51%|█████     | 148/291 [00:02<00:02, 66.95it/s]
Loading 0:  54%|█████▍    | 157/291 [00:03<00:02, 63.70it/s]
Loading 0:  57%|█████▋    | 166/291 [00:03<00:01, 63.57it/s]
Loading 0:  60%|█████▉    | 174/291 [00:03<00:02, 45.27it/s]
Loading 0:  62%|██████▏   | 180/291 [00:03<00:02, 45.34it/s]
Loading 0:  64%|██████▍   | 187/291 [00:04<00:06, 16.17it/s]
Loading 0:  69%|██████▉   | 202/291 [00:04<00:03, 26.41it/s]
Loading 0:  73%|███████▎  | 211/291 [00:05<00:02, 32.45it/s]
Loading 0:  76%|███████▌  | 220/291 [00:05<00:01, 38.97it/s]
Loading 0:  79%|███████▉  | 231/291 [00:05<00:01, 49.36it/s]
Loading 0:  83%|████████▎ | 241/291 [00:05<00:00, 54.14it/s]
Loading 0:  88%|████████▊ | 256/291 [00:05<00:00, 67.53it/s]
Loading 0:  92%|█████████▏| 267/291 [00:05<00:00, 76.06it/s]
Loading 0:  95%|█████████▌| 277/291 [00:05<00:00, 73.17it/s]
Loading 0:  98%|█████████▊| 286/291 [00:05<00:00, 75.24it/s]
                                                            
Job rinen0721-llama0914-1426-v3-mkmlizer completed after 83.99s with status: succeeded
Stopping job with name rinen0721-llama0914-1426-v3-mkmlizer
Pipeline stage MKMLizer completed in 86.56s
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 rinen0721-llama0914-1426-v3
Waiting for inference service rinen0721-llama0914-1426-v3 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 rinen0721-llama0914-1426-v3 ready after 171.54380536079407s
Pipeline stage MKMLDeployer completed in 173.33s
run pipeline stage %s
Failed to get response for submission blend_hokok_2024-09-09: ('http://neversleep-noromaid-v0-8068-v150-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '')
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1637303829193115s
Received healthy response to inference request in 1.9338452816009521s
Received healthy response to inference request in 1.642390489578247s
Received healthy response to inference request in 1.872159481048584s
Received healthy response to inference request in 1.5410540103912354s
5 requests
0 failed requests
5th percentile: 1.5613213062286377
10th percentile: 1.5815886020660401
20th percentile: 1.6221231937408447
30th percentile: 1.6883442878723145
40th percentile: 1.7802518844604491
50th percentile: 1.872159481048584
60th percentile: 1.8968338012695312
70th percentile: 1.9215081214904786
80th percentile: 1.9798223018646242
90th percentile: 2.071776342391968
95th percentile: 2.1177533626556397
99th percentile: 2.154534978866577
mean time: 1.830635929107666
Pipeline stage StressChecker completed in 12.68s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%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
Pipeline stage TriggerMKMLProfilingPipeline completed in 8.80s
Shutdown handler de-registered
rinen0721-llama0914-1426_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 0.12s
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 rinen0721-llama0914-1426-v3-profiler
Waiting for inference service rinen0721-llama0914-1426-v3-profiler to be ready
Inference service rinen0721-llama0914-1426-v3-profiler ready after 170.42997765541077s
Pipeline stage MKMLProfilerDeployer completed in 175.37s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rinen0721-llama0914-1426-v3-profiler-predictor-00001-deplo9xkhd:/code/chaiverse_profiler_1726318854 --namespace tenant-chaiml-guanaco
kubectl exec -it rinen0721-llama0914-1426-v3-profiler-predictor-00001-deplo9xkhd --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726318854 && 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_1726318854/summary.json'
kubectl exec -it rinen0721-llama0914-1426-v3-profiler-predictor-00001-deplo9xkhd --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726318854/summary.json'
Pipeline stage MKMLProfilerRunner completed in 844.38s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rinen0721-llama0914-1426-v3-profiler is running
Tearing down inference service rinen0721-llama0914-1426-v3-profiler
Service rinen0721-llama0914-1426-v3-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.99s
Shutdown handler de-registered
rinen0721-llama0914-1426_v3 status is now inactive due to auto deactivation removed underperforming models
rinen0721-llama0914-1426_v3 status is now torndown due to DeploymentManager action