submission_id: jic062-nemo-v1-7_v2
developer_uid: chace9580
best_of: 8
celo_rating: 1264.2
display_name: jic062-nemo-v1-7_v2
family_friendly_score: 0.0
formatter: {'memory_template': '[INST]system\n{memory}[/INST]\n', 'prompt_template': '[INST]user\n{prompt}[/INST]\n', 'bot_template': '[INST]assistant\n{bot_name}: {message}[/INST]\n', 'user_template': '[INST]user\n{user_name}: {message}[/INST]\n', 'response_template': '[INST]assistant\n{bot_name}:', 'truncate_by_message': False}
generation_params: {'temperature': 1.0, 'top_p': 0.9, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '/s', '[/INST]'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}
gpu_counts: {'NVIDIA RTX A5000': 1}
is_internal_developer: False
language_model: jic062/Nemo-v1.7
latencies: [{'batch_size': 1, 'throughput': 0.6171053110388256, 'latency_mean': 1.6204071688652038, 'latency_p50': 1.6213077306747437, 'latency_p90': 1.7945618629455566}, {'batch_size': 3, 'throughput': 1.0937219762613377, 'latency_mean': 2.7328527879714968, 'latency_p50': 2.732895851135254, 'latency_p90': 3.039970302581787}, {'batch_size': 5, 'throughput': 1.2525322932442662, 'latency_mean': 3.985844533443451, 'latency_p50': 4.001837968826294, 'latency_p90': 4.459808611869812}, {'batch_size': 6, 'throughput': 1.2790709150982504, 'latency_mean': 4.669111788272858, 'latency_p50': 4.722796201705933, 'latency_p90': 5.190320992469787}, {'batch_size': 8, 'throughput': 1.2552694206777033, 'latency_mean': 6.33137892127037, 'latency_p50': 6.318492770195007, 'latency_p90': 7.291156864166259}, {'batch_size': 10, 'throughput': 1.2055955908453857, 'latency_mean': 8.239238364696503, 'latency_p50': 8.387887954711914, 'latency_p90': 9.24934742450714}]
max_input_tokens: 1024
max_output_tokens: 64
model_architecture: MistralForCausalLM
model_group: jic062/Nemo-v1.7
model_name: jic062-nemo-v1-7_v2
model_num_parameters: 12772070400.0
model_repo: jic062/Nemo-v1.7
model_size: 13B
num_battles: 22684
num_wins: 12129
ranking_group: single
status: torndown
submission_type: basic
throughput_3p7s: 1.23
timestamp: 2024-09-24T03:25:55+00:00
us_pacific_date: 2024-09-23
win_ratio: 0.5346940574854523
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 jic062-nemo-v1-7-v2-mkmlizer
Waiting for job on jic062-nemo-v1-7-v2-mkmlizer to finish
jic062-nemo-v1-7-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-nemo-v1-7-v2-mkmlizer: ║ _____ __ __ ║
jic062-nemo-v1-7-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
jic062-nemo-v1-7-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
jic062-nemo-v1-7-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
jic062-nemo-v1-7-v2-mkmlizer: ║ /___/ ║
jic062-nemo-v1-7-v2-mkmlizer: ║ ║
jic062-nemo-v1-7-v2-mkmlizer: ║ Version: 0.10.1 ║
jic062-nemo-v1-7-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
jic062-nemo-v1-7-v2-mkmlizer: ║ https://mk1.ai ║
jic062-nemo-v1-7-v2-mkmlizer: ║ ║
jic062-nemo-v1-7-v2-mkmlizer: ║ The license key for the current software has been verified as ║
jic062-nemo-v1-7-v2-mkmlizer: ║ belonging to: ║
jic062-nemo-v1-7-v2-mkmlizer: ║ ║
jic062-nemo-v1-7-v2-mkmlizer: ║ Chai Research Corp. ║
jic062-nemo-v1-7-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
jic062-nemo-v1-7-v2-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
jic062-nemo-v1-7-v2-mkmlizer: ║ ║
jic062-nemo-v1-7-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Connection pool is full, discarding connection: %s. Connection pool size: %s
jic062-nemo-v1-7-v2-mkmlizer: Downloaded to shared memory in 45.917s
jic062-nemo-v1-7-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpfut7kfy0, device:0
jic062-nemo-v1-7-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Failed to get response for submission blend_negek_2024-09-20: ('http://zonemercy-lexical-nemov8-5966-v9-predictor.tenant-chaiml-guanaco.k2.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:47864->127.0.0.1:8080: read: connection reset by peer\n')
jic062-nemo-v1-7-v2-mkmlizer: quantized model in 35.654s
jic062-nemo-v1-7-v2-mkmlizer: Processed model jic062/Nemo-v1.7 in 81.572s
jic062-nemo-v1-7-v2-mkmlizer: creating bucket guanaco-mkml-models
jic062-nemo-v1-7-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-nemo-v1-7-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-nemo-v1-7-v2
jic062-nemo-v1-7-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-nemo-v1-7-v2/config.json
jic062-nemo-v1-7-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-nemo-v1-7-v2/special_tokens_map.json
jic062-nemo-v1-7-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-nemo-v1-7-v2/tokenizer_config.json
jic062-nemo-v1-7-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-nemo-v1-7-v2/tokenizer.json
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
Job jic062-nemo-v1-7-v2-mkmlizer completed after 108.9s with status: succeeded
Stopping job with name jic062-nemo-v1-7-v2-mkmlizer
Pipeline stage MKMLizer completed in 109.94s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.11s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service jic062-nemo-v1-7-v2
Waiting for inference service jic062-nemo-v1-7-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
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 jic062-nemo-v1-7-v2 ready after 224.03646183013916s
Pipeline stage MKMLDeployer completed in 224.51s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.501481533050537s
Received healthy response to inference request in 2.5598556995391846s
Received healthy response to inference request in 1.773000717163086s
Received healthy response to inference request in 1.9527854919433594s
Received healthy response to inference request in 1.797614336013794s
5 requests
0 failed requests
5th percentile: 1.7779234409332276
10th percentile: 1.782846164703369
20th percentile: 1.7926916122436523
30th percentile: 1.828648567199707
40th percentile: 1.8907170295715332
50th percentile: 1.9527854919433594
60th percentile: 2.1722639083862303
70th percentile: 2.3917423248291017
80th percentile: 2.513156366348267
90th percentile: 2.5365060329437257
95th percentile: 2.548180866241455
99th percentile: 2.5575207328796385
mean time: 2.116947555541992
Pipeline stage StressChecker completed in 11.85s
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 6.91s
Shutdown handler de-registered
jic062-nemo-v1-7_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 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service jic062-nemo-v1-7-v2-profiler
Waiting for inference service jic062-nemo-v1-7-v2-profiler to be ready
Inference service jic062-nemo-v1-7-v2-profiler ready after 220.50219678878784s
Pipeline stage MKMLProfilerDeployer completed in 220.86s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/jic062-nemo-v1-7-v2-profiler-predictor-00001-deployment-57kvjrm:/code/chaiverse_profiler_1727148982 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-nemo-v1-7-v2-profiler-predictor-00001-deployment-57kvjrm --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1727148982 && 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_1727148982/summary.json'
kubectl exec -it jic062-nemo-v1-7-v2-profiler-predictor-00001-deployment-57kvjrm --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1727148982/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1155.16s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-nemo-v1-7-v2-profiler is running
Tearing down inference service jic062-nemo-v1-7-v2-profiler
Service jic062-nemo-v1-7-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.02s
Shutdown handler de-registered
jic062-nemo-v1-7_v2 status is now inactive due to auto deactivation removed underperforming models
run pipeline %s
admin requested tearing down of jic062-nemo-v1-7_v2
run pipeline stage %s
Shutdown handler not registered because Python interpreter is not running in the main thread
Running pipeline stage MKMLDeleter
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLDeleter
%s, retrying in %s seconds...
%s, retrying in %s seconds...
clean up pipeline due to error=TeardownError("module 'kubernetes.config' has no attribute 'load_kube_config'")
Shutdown handler de-registered
jic062-nemo-v1-7_v2 status is now torndown due to DeploymentManager action