submission_id: rica40325-feedback-2000_v3
developer_uid: rica40325
best_of: 16
celo_rating: 1257.26
display_name: rica40325-feedback-2000_v1
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': 1.15, 'top_p': 0.95, 'min_p': 0.1, 'top_k': 80, '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: rica40325/feedback_2000
latencies: [{'batch_size': 1, 'throughput': 0.9043414758504397, 'latency_mean': 1.1056831681728363, 'latency_p50': 1.111937165260315, 'latency_p90': 1.2449367761611938}, {'batch_size': 4, 'throughput': 1.8003848904416424, 'latency_mean': 2.213484400510788, 'latency_p50': 2.2333524227142334, 'latency_p90': 2.4766613721847532}, {'batch_size': 5, 'throughput': 1.9025321071094754, 'latency_mean': 2.608798418045044, 'latency_p50': 2.60060977935791, 'latency_p90': 2.927461838722229}, {'batch_size': 8, 'throughput': 2.002715880305393, 'latency_mean': 3.9703422737121583, 'latency_p50': 3.9733803272247314, 'latency_p90': 4.441733956336975}, {'batch_size': 10, 'throughput': 2.053459153244199, 'latency_mean': 4.816543097496033, 'latency_p50': 4.833685755729675, 'latency_p90': 5.53416678905487}, {'batch_size': 12, 'throughput': 2.0286851050587558, 'latency_mean': 5.84325400352478, 'latency_p50': 5.853449702262878, 'latency_p90': 6.770500636100769}, {'batch_size': 15, 'throughput': 2.0439834925939477, 'latency_mean': 7.200194363594055, 'latency_p50': 7.317592978477478, 'latency_p90': 7.892062473297119}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: rica40325/feedback_2000
model_name: rica40325-feedback-2000_v1
model_num_parameters: 8030261248.0
model_repo: rica40325/feedback_2000
model_size: 8B
num_battles: 12590
num_wins: 6576
ranking_group: single
status: torndown
submission_type: basic
throughput_3p7s: 2.0
timestamp: 2024-09-07T17:03:07+00:00
us_pacific_date: 2024-09-07
win_ratio: 0.5223193010325655
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 rica40325-feedback-2000-v3-mkmlizer
Waiting for job on rica40325-feedback-2000-v3-mkmlizer to finish
rica40325-feedback-2000-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-feedback-2000-v3-mkmlizer: ║ _____ __ __ ║
rica40325-feedback-2000-v3-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
rica40325-feedback-2000-v3-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
rica40325-feedback-2000-v3-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
rica40325-feedback-2000-v3-mkmlizer: ║ /___/ ║
rica40325-feedback-2000-v3-mkmlizer: ║ ║
rica40325-feedback-2000-v3-mkmlizer: ║ Version: 0.10.1 ║
rica40325-feedback-2000-v3-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
rica40325-feedback-2000-v3-mkmlizer: ║ https://mk1.ai ║
rica40325-feedback-2000-v3-mkmlizer: ║ ║
rica40325-feedback-2000-v3-mkmlizer: ║ The license key for the current software has been verified as ║
rica40325-feedback-2000-v3-mkmlizer: ║ belonging to: ║
rica40325-feedback-2000-v3-mkmlizer: ║ ║
rica40325-feedback-2000-v3-mkmlizer: ║ Chai Research Corp. ║
rica40325-feedback-2000-v3-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
rica40325-feedback-2000-v3-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
rica40325-feedback-2000-v3-mkmlizer: ║ ║
rica40325-feedback-2000-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rica40325-feedback-2000-v3-mkmlizer: Downloaded to shared memory in 41.412s
rica40325-feedback-2000-v3-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp0qopydoi, device:0
rica40325-feedback-2000-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Failed to get response for submission mistralai-mixtral-8x7b_3473_v131: ('http://chaiml-llama-8b-pairwis-8189-v19-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:59804->127.0.0.1:8080: read: connection reset by peer\n')
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
rica40325-feedback-2000-v3-mkmlizer: quantized model in 29.185s
rica40325-feedback-2000-v3-mkmlizer: Processed model rica40325/feedback_2000 in 70.597s
rica40325-feedback-2000-v3-mkmlizer: creating bucket guanaco-mkml-models
rica40325-feedback-2000-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-feedback-2000-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-feedback-2000-v3
rica40325-feedback-2000-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-feedback-2000-v3/config.json
rica40325-feedback-2000-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-feedback-2000-v3/special_tokens_map.json
rica40325-feedback-2000-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-feedback-2000-v3/tokenizer_config.json
rica40325-feedback-2000-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-feedback-2000-v3/tokenizer.json
rica40325-feedback-2000-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-feedback-2000-v3/flywheel_model.0.safetensors
rica40325-feedback-2000-v3-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:10, 27.50it/s] Loading 0: 4%|▍ | 12/291 [00:00<00:07, 36.68it/s] Loading 0: 5%|▌ | 16/291 [00:00<00:08, 33.76it/s] Loading 0: 7%|▋ | 21/291 [00:00<00:07, 35.46it/s] Loading 0: 9%|▊ | 25/291 [00:00<00:07, 33.89it/s] Loading 0: 10%|█ | 30/291 [00:00<00:06, 37.58it/s] Loading 0: 12%|█▏ | 34/291 [00:01<00:09, 26.11it/s] Loading 0: 13%|█▎ | 38/291 [00:01<00:09, 27.34it/s] Loading 0: 14%|█▍ | 42/291 [00:01<00:09, 25.44it/s] Loading 0: 16%|█▋ | 48/291 [00:01<00:07, 30.81it/s] Loading 0: 18%|█▊ | 52/291 [00:01<00:07, 30.31it/s] Loading 0: 20%|█▉ | 57/291 [00:01<00:07, 32.78it/s] Loading 0: 21%|██ | 61/291 [00:01<00:07, 31.90it/s] Loading 0: 23%|██▎ | 66/291 [00:02<00:06, 34.69it/s] Loading 0: 24%|██▍ | 70/291 [00:02<00:06, 32.83it/s] Loading 0: 25%|██▌ | 74/291 [00:02<00:06, 33.24it/s] Loading 0: 27%|██▋ | 78/291 [00:02<00:06, 33.30it/s] Loading 0: 28%|██▊ | 82/291 [00:02<00:08, 23.52it/s] Loading 0: 29%|██▉ | 85/291 [00:02<00:08, 24.45it/s] Loading 0: 31%|███ | 90/291 [00:02<00:07, 27.87it/s] Loading 0: 32%|███▏ | 94/291 [00:03<00:07, 28.12it/s] Loading 0: 34%|███▍ | 99/291 [00:03<00:06, 31.60it/s] Loading 0: 35%|███▌ | 103/291 [00:03<00:06, 31.26it/s] Loading 0: 37%|███▋ | 108/291 [00:03<00:05, 33.42it/s] Loading 0: 38%|███▊ | 112/291 [00:03<00:05, 32.13it/s] Loading 0: 40%|███▉ | 116/291 [00:03<00:05, 32.55it/s] Loading 0: 42%|████▏ | 122/291 [00:03<00:04, 36.90it/s] Loading 0: 44%|████▎ | 127/291 [00:04<00:04, 34.75it/s] Loading 0: 46%|████▌ | 133/291 [00:04<00:05, 28.86it/s] Loading 0: 47%|████▋ | 137/291 [00:04<00:05, 29.29it/s] Loading 0: 48%|████▊ | 141/291 [00:04<00:05, 27.85it/s] Loading 0: 51%|█████ | 147/291 [00:04<00:04, 32.85it/s] Loading 0: 52%|█████▏ | 151/291 [00:04<00:04, 32.09it/s] Loading 0: 54%|█████▎ | 156/291 [00:04<00:03, 34.52it/s] Loading 0: 55%|█████▍ | 160/291 [00:05<00:03, 33.17it/s] Loading 0: 57%|█████▋ | 165/291 [00:05<00:03, 35.83it/s] Loading 0: 58%|█████▊ | 169/291 [00:05<00:03, 34.17it/s] Loading 0: 60%|█████▉ | 174/291 [00:05<00:03, 35.95it/s] Loading 0: 61%|██████ | 178/291 [00:05<00:03, 34.05it/s] Loading 0: 63%|██████▎ | 183/291 [00:05<00:02, 37.33it/s] Loading 0: 64%|██████▍ | 187/291 [00:06<00:03, 26.33it/s] Loading 0: 66%|██████▌ | 191/291 [00:06<00:03, 26.76it/s] Loading 0: 67%|██████▋ | 195/291 [00:06<00:03, 25.96it/s] Loading 0: 69%|██████▉ | 201/291 [00:06<00:02, 31.58it/s] Loading 0: 70%|███████ | 205/291 [00:06<00:02, 31.51it/s] Loading 0: 72%|███████▏ | 210/291 [00:06<00:02, 34.00it/s] Loading 0: 74%|███████▎ | 214/291 [00:06<00:02, 32.79it/s] Loading 0: 75%|███████▌ | 219/291 [00:06<00:02, 35.19it/s] Loading 0: 77%|███████▋ | 223/291 [00:07<00:02, 33.28it/s] Loading 0: 78%|███████▊ | 227/291 [00:07<00:02, 31.95it/s] Loading 0: 79%|███████▉ | 231/291 [00:07<00:01, 32.11it/s] Loading 0: 81%|████████ | 235/291 [00:07<00:02, 24.63it/s] Loading 0: 82%|████████▏ | 239/291 [00:07<00:02, 24.85it/s] Loading 0: 85%|████████▍ | 246/291 [00:07<00:01, 32.74it/s] Loading 0: 86%|████████▌ | 250/291 [00:08<00:01, 32.15it/s] Loading 0: 88%|████████▊ | 255/291 [00:08<00:01, 32.74it/s] Loading 0: 89%|████████▉ | 259/291 [00:08<00:01, 31.90it/s] Loading 0: 91%|█████████ | 264/291 [00:08<00:00, 33.28it/s] Loading 0: 92%|█████████▏| 268/291 [00:08<00:00, 32.20it/s] Loading 0: 94%|█████████▍| 273/291 [00:08<00:00, 35.05it/s] Loading 0: 95%|█████████▌| 277/291 [00:08<00:00, 33.78it/s] Loading 0: 97%|█████████▋| 281/291 [00:08<00:00, 33.50it/s] Loading 0: 98%|█████████▊| 286/291 [00:14<00:01, 2.60it/s] Loading 0: 99%|█████████▉| 289/291 [00:14<00:00, 3.23it/s]
Job rica40325-feedback-2000-v3-mkmlizer completed after 148.63s with status: succeeded
Stopping job with name rica40325-feedback-2000-v3-mkmlizer
Pipeline stage MKMLizer completed in 150.51s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rica40325-feedback-2000-v3
Waiting for inference service rica40325-feedback-2000-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
Failed to get response for submission blend_susol_2024-08-22: ('http://mistralai-mixtral-8x7b-3473-v130-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:47506->127.0.0.1:8080: read: connection reset by peer\n')
Inference service rica40325-feedback-2000-v3 ready after 150.606360912323s
Pipeline stage MKMLDeployer completed in 151.77s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0180583000183105s
Received healthy response to inference request in 1.4236729145050049s
Received healthy response to inference request in 1.7954773902893066s
Received healthy response to inference request in 2.2551510334014893s
Received healthy response to inference request in 1.3893990516662598s
5 requests
0 failed requests
5th percentile: 1.3962538242340088
10th percentile: 1.4031085968017578
20th percentile: 1.4168181419372559
30th percentile: 1.4980338096618653
40th percentile: 1.6467555999755858
50th percentile: 1.7954773902893066
60th percentile: 1.8845097541809082
70th percentile: 1.9735421180725097
80th percentile: 2.0654768466949465
90th percentile: 2.160313940048218
95th percentile: 2.2077324867248533
99th percentile: 2.245667324066162
mean time: 1.7763517379760743
Pipeline stage StressChecker completed in 9.56s
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 4.50s
Shutdown handler de-registered
rica40325-feedback-2000_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 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-feedback-2000-v3-profiler
Waiting for inference service rica40325-feedback-2000-v3-profiler to be ready
Inference service rica40325-feedback-2000-v3-profiler ready after 150.4503939151764s
Pipeline stage MKMLProfilerDeployer completed in 150.83s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-feedback-2000-v3-profiler-predictor-00001-deploycvhx6:/code/chaiverse_profiler_1725729095 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-feedback-2000-v3-profiler-predictor-00001-deploycvhx6 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725729095 && 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_1725729095/summary.json'
kubectl exec -it rica40325-feedback-2000-v3-profiler-predictor-00001-deploycvhx6 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725729095/summary.json'
Pipeline stage MKMLProfilerRunner completed in 839.22s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-feedback-2000-v3-profiler is running
Tearing down inference service rica40325-feedback-2000-v3-profiler
Service rica40325-feedback-2000-v3-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.80s
Shutdown handler de-registered
rica40325-feedback-2000_v3 status is now inactive due to auto deactivation removed underperforming models
rica40325-feedback-2000_v3 status is now torndown due to DeploymentManager action
rica40325-feedback-2000_v3 status is now torndown due to DeploymentManager action