submission_id: epiculous-violet-twiligh_4666_v1
developer_uid: Epiculous
best_of: 8
celo_rating: 1251.81
display_name: Violet_Twilight-v0-2-High-T
family_friendly_score: 0.0
formatter: {'memory_template': '<|im_start|>system\n{memory}<|im_end|>\n', 'prompt_template': '<|im_start|>user\n{prompt}<|im_end|>\n', 'bot_template': '<|im_start|>assistant\n{bot_name}: {message}<|im_end|>\n', 'user_template': '<|im_start|>user\n{user_name}: {message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n{bot_name}:', 'truncate_by_message': False}
generation_params: {'temperature': 5.0, 'top_p': 1.0, 'min_p': 0.55, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}
gpu_counts: {'NVIDIA RTX A5000': 1}
is_internal_developer: False
language_model: Epiculous/Violet_Twilight-v0.2
latencies: [{'batch_size': 1, 'throughput': 0.6992338461227252, 'latency_mean': 1.4300464904308319, 'latency_p50': 1.4278417825698853, 'latency_p90': 1.5818140029907226}, {'batch_size': 3, 'throughput': 1.339085932634582, 'latency_mean': 2.234408733844757, 'latency_p50': 2.235610246658325, 'latency_p90': 2.479797196388245}, {'batch_size': 5, 'throughput': 1.6022799501676628, 'latency_mean': 3.1062233698368074, 'latency_p50': 3.116877317428589, 'latency_p90': 3.475067162513733}, {'batch_size': 6, 'throughput': 1.6375972830755312, 'latency_mean': 3.6470706152915953, 'latency_p50': 3.700829029083252, 'latency_p90': 4.085373497009277}, {'batch_size': 8, 'throughput': 1.6091191603489723, 'latency_mean': 4.932424457073211, 'latency_p50': 4.929110288619995, 'latency_p90': 5.538343667984009}, {'batch_size': 10, 'throughput': 1.5563483051197404, 'latency_mean': 6.38240587592125, 'latency_p50': 6.4279820919036865, 'latency_p90': 7.266347503662109}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: MistralForCausalLM
model_group: Epiculous/Violet_Twiligh
model_name: Violet_Twilight-v0-2-High-T
model_num_parameters: 12772070400.0
model_repo: Epiculous/Violet_Twilight-v0.2
model_size: 13B
num_battles: 11927
num_wins: 6146
ranking_group: single
status: torndown
submission_type: basic
throughput_3p7s: 1.64
timestamp: 2024-09-12T19:46:33+00:00
us_pacific_date: 2024-09-12
win_ratio: 0.5153014169531316
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 epiculous-violet-twiligh-4666-v1-mkmlizer
Waiting for job on epiculous-violet-twiligh-4666-v1-mkmlizer to finish
epiculous-violet-twiligh-4666-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
epiculous-violet-twiligh-4666-v1-mkmlizer: ║     _____            __           __                                ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║       /___/                                                         ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║                                                                     ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║  https://mk1.ai                                                     ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║                                                                     ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║  belonging to:                                                      ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║                                                                     ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║  Chai Research Corp.                                                ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ║                                                                     ║
epiculous-violet-twiligh-4666-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Connection pool is full, discarding connection: %s. Connection pool size: %s
epiculous-violet-twiligh-4666-v1-mkmlizer: Downloaded to shared memory in 37.913s
epiculous-violet-twiligh-4666-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpd4eavjtj, device:0
epiculous-violet-twiligh-4666-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
epiculous-violet-twiligh-4666-v1-mkmlizer: quantized model in 35.854s
epiculous-violet-twiligh-4666-v1-mkmlizer: Processed model Epiculous/Violet_Twilight-v0.2 in 73.768s
epiculous-violet-twiligh-4666-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
epiculous-violet-twiligh-4666-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/epiculous-violet-twiligh-4666-v1
epiculous-violet-twiligh-4666-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/epiculous-violet-twiligh-4666-v1/config.json
epiculous-violet-twiligh-4666-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/epiculous-violet-twiligh-4666-v1/special_tokens_map.json
epiculous-violet-twiligh-4666-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/epiculous-violet-twiligh-4666-v1/tokenizer_config.json
epiculous-violet-twiligh-4666-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/epiculous-violet-twiligh-4666-v1/tokenizer.json
epiculous-violet-twiligh-4666-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/epiculous-violet-twiligh-4666-v1/flywheel_model.0.safetensors
Job epiculous-violet-twiligh-4666-v1-mkmlizer completed after 105.34s with status: succeeded
Stopping job with name epiculous-violet-twiligh-4666-v1-mkmlizer
Pipeline stage MKMLizer completed in 106.26s
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 epiculous-violet-twiligh-4666-v1
Waiting for inference service epiculous-violet-twiligh-4666-v1 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
Inference service epiculous-violet-twiligh-4666-v1 ready after 170.55149269104004s
Pipeline stage MKMLDeployer completed in 170.95s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.402426242828369s
Received healthy response to inference request in 3.015423059463501s
Received healthy response to inference request in 1.8561224937438965s
Received healthy response to inference request in 2.4261341094970703s
Received healthy response to inference request in 1.791083574295044s
5 requests
0 failed requests
5th percentile: 1.8040913581848144
10th percentile: 1.817099142074585
20th percentile: 1.843114709854126
30th percentile: 1.965383243560791
40th percentile: 2.18390474319458
50th percentile: 2.402426242828369
60th percentile: 2.4119093894958494
70th percentile: 2.42139253616333
80th percentile: 2.5439918994903565
90th percentile: 2.7797074794769285
95th percentile: 2.8975652694702148
99th percentile: 2.991851501464844
mean time: 2.2982378959655763
Pipeline stage StressChecker completed in 12.38s
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 5.69s
Shutdown handler de-registered
epiculous-violet-twiligh_4666_v1 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 0.11s
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 epiculous-violet-twiligh-4666-v1-profiler
Waiting for inference service epiculous-violet-twiligh-4666-v1-profiler to be ready
Inference service epiculous-violet-twiligh-4666-v1-profiler ready after 170.4218783378601s
Pipeline stage MKMLProfilerDeployer completed in 170.82s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/epiculous-violet-twi2c231ccb0ed8da3270d896622355d183-deplo4gkrt:/code/chaiverse_profiler_1726170910 --namespace tenant-chaiml-guanaco
kubectl exec -it epiculous-violet-twi2c231ccb0ed8da3270d896622355d183-deplo4gkrt --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726170910 && 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 512 --output_tokens 64 --summary /code/chaiverse_profiler_1726170910/summary.json'
kubectl exec -it epiculous-violet-twi2c231ccb0ed8da3270d896622355d183-deplo4gkrt --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726170910/summary.json'
Pipeline stage MKMLProfilerRunner completed in 940.73s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service epiculous-violet-twiligh-4666-v1-profiler is running
Tearing down inference service epiculous-violet-twiligh-4666-v1-profiler
Service epiculous-violet-twiligh-4666-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 3.51s
Shutdown handler de-registered
epiculous-violet-twiligh_4666_v1 status is now inactive due to auto deactivation removed underperforming models
epiculous-violet-twiligh_4666_v1 status is now torndown due to DeploymentManager action