jic062-instruct-v14-v15

developer_uid: chace9580

submission_id: jic062-instruct-v14-v15_v5

model_name: jic062-instruct-v14-v15_v5

model_group: jic062/instruct_v14_v15

status: torndown

timestamp: 2024-09-15T05:01:17+00:00

num_battles: 10794

num_wins: 5330

celo_rating: 1244.46

family_friendly_score: 0.0

submission_type: basic

model_repo: jic062/instruct_v14_v15

model_architecture: LlamaForCausalLM

model_num_parameters: 8030261248.0

best_of: 16

max_input_tokens: 512

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.8994932143442248, 'latency_mean': 1.1116758275032044, 'latency_p50': 1.1186515092849731, 'latency_p90': 1.2284764289855956}, {'batch_size': 3, 'throughput': 1.600027328957998, 'latency_mean': 1.8735782754421235, 'latency_p50': 1.8842507600784302, 'latency_p90': 2.0609854459762573}, {'batch_size': 5, 'throughput': 1.7582448206921377, 'latency_mean': 2.834015454053879, 'latency_p50': 2.8357573747634888, 'latency_p90': 3.2169594049453734}, {'batch_size': 6, 'throughput': 1.7537561760336173, 'latency_mean': 3.407741417884827, 'latency_p50': 3.4134252071380615, 'latency_p90': 3.847043490409851}, {'batch_size': 8, 'throughput': 1.7414010671536202, 'latency_mean': 4.55832869052887, 'latency_p50': 4.545124053955078, 'latency_p90': 5.116367959976197}, {'batch_size': 10, 'throughput': 1.765101543352641, 'latency_mean': 5.607985202074051, 'latency_p50': 5.595300793647766, 'latency_p90': 6.4878675699234005}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: jic062-instruct-v14-v15_v5

is_internal_developer: False

language_model: jic062/instruct_v14_v15

model_size: 8B

ranking_group: single

throughput_3p7s: 1.76

us_pacific_date: 2024-09-14

win_ratio: 0.493792847878451

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', '<|end_of_text|>', '|eot_id|'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

formatter: {'memory_template': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}:', 'truncate_by_message': False}

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 jic062-instruct-v14-v15-v5-mkmlizer
Waiting for job on jic062-instruct-v14-v15-v5-mkmlizer to finish
jic062-instruct-v14-v15-v5-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-instruct-v14-v15-v5-mkmlizer: ║     _____            __           __                                ║
jic062-instruct-v14-v15-v5-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-instruct-v14-v15-v5-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-instruct-v14-v15-v5-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-instruct-v14-v15-v5-mkmlizer: ║       /___/                                                         ║
jic062-instruct-v14-v15-v5-mkmlizer: ║                                                                     ║
Connection pool is full, discarding connection: %s. Connection pool size: %s
jic062-instruct-v14-v15-v5-mkmlizer: ║  Version: 0.10.1                                                    ║
jic062-instruct-v14-v15-v5-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-instruct-v14-v15-v5-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-instruct-v14-v15-v5-mkmlizer: ║                                                                     ║
jic062-instruct-v14-v15-v5-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-instruct-v14-v15-v5-mkmlizer: ║  belonging to:                                                      ║
jic062-instruct-v14-v15-v5-mkmlizer: ║                                                                     ║
Connection pool is full, discarding connection: %s. Connection pool size: %s
jic062-instruct-v14-v15-v5-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-instruct-v14-v15-v5-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-instruct-v14-v15-v5-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-instruct-v14-v15-v5-mkmlizer: ║                                                                     ║
jic062-instruct-v14-v15-v5-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
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
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
jic062-instruct-v14-v15-v5-mkmlizer: Downloaded to shared memory in 36.507s
jic062-instruct-v14-v15-v5-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpkke8tszr, device:0
jic062-instruct-v14-v15-v5-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-instruct-v14-v15-v5-mkmlizer: quantized model in 25.995s
jic062-instruct-v14-v15-v5-mkmlizer: Processed model jic062/instruct_v14_v15 in 62.503s
jic062-instruct-v14-v15-v5-mkmlizer: creating bucket guanaco-mkml-models
jic062-instruct-v14-v15-v5-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-instruct-v14-v15-v5-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-instruct-v14-v15-v5
jic062-instruct-v14-v15-v5-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-instruct-v14-v15-v5/special_tokens_map.json
jic062-instruct-v14-v15-v5-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-instruct-v14-v15-v5/config.json
jic062-instruct-v14-v15-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-instruct-v14-v15-v5/tokenizer_config.json
jic062-instruct-v14-v15-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-instruct-v14-v15-v5/tokenizer.json
jic062-instruct-v14-v15-v5-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-instruct-v14-v15-v5/flywheel_model.0.safetensors
jic062-instruct-v14-v15-v5-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 50.46it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:03, 70.26it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:03, 76.50it/s]
Loading 0:  12%|█▏        | 34/291 [00:00<00:03, 78.11it/s]
Loading 0:  15%|█▍        | 43/291 [00:00<00:03, 79.56it/s]
Loading 0:  18%|█▊        | 52/291 [00:00<00:02, 82.55it/s]
Loading 0:  21%|██        | 61/291 [00:00<00:02, 77.88it/s]
Loading 0:  26%|██▌       | 76/291 [00:00<00:02, 86.68it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:08, 22.92it/s]
Loading 0:  33%|███▎      | 95/291 [00:02<00:06, 29.90it/s]
Loading 0:  36%|███▋      | 106/291 [00:02<00:04, 37.10it/s]
Loading 0:  40%|███▉      | 115/291 [00:02<00:04, 43.68it/s]
Loading 0:  43%|████▎     | 124/291 [00:02<00:03, 50.13it/s]
Loading 0:  46%|████▌     | 133/291 [00:02<00:02, 54.60it/s]
Loading 0:  49%|████▉     | 143/291 [00:02<00:02, 63.45it/s]
Loading 0:  52%|█████▏    | 152/291 [00:02<00:02, 65.68it/s]
Loading 0:  55%|█████▍    | 160/291 [00:03<00:01, 67.58it/s]
Loading 0:  58%|█████▊    | 169/291 [00:03<00:01, 69.51it/s]
Loading 0:  61%|██████    | 178/291 [00:03<00:01, 73.67it/s]
Loading 0:  64%|██████▍   | 187/291 [00:04<00:04, 21.91it/s]
Loading 0:  67%|██████▋   | 196/291 [00:04<00:03, 28.24it/s]
Loading 0:  70%|███████   | 205/291 [00:04<00:02, 35.07it/s]
Loading 0:  74%|███████▎  | 214/291 [00:04<00:01, 42.69it/s]
Loading 0:  77%|███████▋  | 223/291 [00:04<00:01, 49.62it/s]
Loading 0:  80%|███████▉  | 232/291 [00:04<00:01, 53.21it/s]
Loading 0:  83%|████████▎ | 241/291 [00:05<00:00, 56.47it/s]
Loading 0:  86%|████████▌ | 250/291 [00:05<00:00, 61.83it/s]
Loading 0:  89%|████████▉ | 259/291 [00:05<00:00, 64.29it/s]
Loading 0:  92%|█████████▏| 268/291 [00:05<00:00, 66.08it/s]
Loading 0:  96%|█████████▌| 278/291 [00:05<00:00, 74.13it/s]
Loading 0:  99%|█████████▊| 287/291 [00:05<00:00, 41.89it/s]
                                                            
Job jic062-instruct-v14-v15-v5-mkmlizer completed after 85.64s with status: succeeded
Stopping job with name jic062-instruct-v14-v15-v5-mkmlizer
Pipeline stage MKMLizer completed in 86.65s
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 jic062-instruct-v14-v15-v5
Waiting for inference service jic062-instruct-v14-v15-v5 to be ready
Failed to get response for submission blend_sanen_2024-09-09: ('http://rica40325-lora32-third-step-v4-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:59774->127.0.0.1:8080: read: connection reset by peer\n')
Inference service jic062-instruct-v14-v15-v5 ready after 160.5797574520111s
Pipeline stage MKMLDeployer completed in 162.66s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.106239080429077s
Received healthy response to inference request in 2.7862703800201416s
Received healthy response to inference request in 2.0209767818450928s
Received healthy response to inference request in 1.726043462753296s
Received healthy response to inference request in 2.88433575630188s
5 requests
0 failed requests
5th percentile: 1.7850301265716553
10th percentile: 1.8440167903900146
20th percentile: 1.9619901180267334
30th percentile: 2.0380292415618895
40th percentile: 2.0721341609954833
50th percentile: 2.106239080429077
60th percentile: 2.3782516002655028
70th percentile: 2.6502641201019284
80th percentile: 2.805883455276489
90th percentile: 2.8451096057891845
95th percentile: 2.8647226810455324
99th percentile: 2.88041314125061
mean time: 2.3047730922698975
Pipeline stage StressChecker completed in 15.63s
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 5.99s
Shutdown handler de-registered
jic062-instruct-v14-v15_v5 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 0.13s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.15s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service jic062-instruct-v14-v15-v5-profiler
Waiting for inference service jic062-instruct-v14-v15-v5-profiler to be ready
Inference service jic062-instruct-v14-v15-v5-profiler ready after 170.41042733192444s
Pipeline stage MKMLProfilerDeployer completed in 170.83s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/jic062-instruct-v14-v15-v5-profiler-predictor-00001-deployxmwkd:/code/chaiverse_profiler_1726376959 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-instruct-v14-v15-v5-profiler-predictor-00001-deployxmwkd --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726376959 && 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_1726376959/summary.json'
kubectl exec -it jic062-instruct-v14-v15-v5-profiler-predictor-00001-deployxmwkd --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726376959/summary.json'
Pipeline stage MKMLProfilerRunner completed in 809.30s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-instruct-v14-v15-v5-profiler is running
Tearing down inference service jic062-instruct-v14-v15-v5-profiler
Service jic062-instruct-v14-v15-v5-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.81s
Shutdown handler de-registered
jic062-instruct-v14-v15_v5 status is now inactive due to auto deactivation removed underperforming models
jic062-instruct-v14-v15_v5 status is now torndown due to DeploymentManager action