developer_uid: rica40325
submission_id: rica40325-9-15-sftdpo_v4
model_name: rica40325-9-15-sftdpo_v1
model_group: rica40325/9-15-sftdpo
status: torndown
timestamp: 2024-09-16T17:35:38+00:00
num_battles: 12014
num_wins: 5922
celo_rating: 1248.44
family_friendly_score: 0.0
submission_type: basic
model_repo: rica40325/9-15-sftdpo
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 8
max_input_tokens: 512
max_output_tokens: 64
latencies: [{'batch_size': 1, 'throughput': 0.9725092838893109, 'latency_mean': 1.0281765007972716, 'latency_p50': 1.02676260471344, 'latency_p90': 1.1433743000030518}, {'batch_size': 5, 'throughput': 2.5490701693052165, 'latency_mean': 1.952574771642685, 'latency_p50': 1.9603886604309082, 'latency_p90': 2.1878897666931154}, {'batch_size': 10, 'throughput': 3.0882112391903505, 'latency_mean': 3.2051947009563446, 'latency_p50': 3.208631753921509, 'latency_p90': 3.5943587541580198}, {'batch_size': 15, 'throughput': 3.2508348077773013, 'latency_mean': 4.559766167402268, 'latency_p50': 4.550343036651611, 'latency_p90': 5.181478071212768}, {'batch_size': 20, 'throughput': 3.3166987195695277, 'latency_mean': 5.937364364862442, 'latency_p50': 5.971385836601257, 'latency_p90': 6.747831106185913}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: rica40325-9-15-sftdpo_v1
is_internal_developer: False
language_model: rica40325/9-15-sftdpo
model_size: 8B
ranking_group: single
throughput_3p7s: 3.2
us_pacific_date: 2024-09-16
win_ratio: 0.4929249209255868
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': 8, 'max_output_tokens': 64}
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}
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 rica40325-9-15-sftdpo-v4-mkmlizer
Waiting for job on rica40325-9-15-sftdpo-v4-mkmlizer to finish
rica40325-9-15-sftdpo-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-9-15-sftdpo-v4-mkmlizer: ║     _____            __           __                                ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║       /___/                                                         ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║                                                                     ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║  Version: 0.10.1                                                    ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║  https://mk1.ai                                                     ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║                                                                     ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║  The license key for the current software has been verified as      ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║  belonging to:                                                      ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║                                                                     ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║  Chai Research Corp.                                                ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rica40325-9-15-sftdpo-v4-mkmlizer: ║                                                                     ║
rica40325-9-15-sftdpo-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rica40325-9-15-sftdpo-v4-mkmlizer: Downloaded to shared memory in 19.784s
rica40325-9-15-sftdpo-v4-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmprchmtyso, device:0
rica40325-9-15-sftdpo-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-9-15-sftdpo-v4-mkmlizer: quantized model in 25.360s
rica40325-9-15-sftdpo-v4-mkmlizer: Processed model rica40325/9-15-sftdpo in 45.144s
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', '')
rica40325-9-15-sftdpo-v4-mkmlizer: creating bucket guanaco-mkml-models
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_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', '')
Connection pool is full, discarding connection: %s. Connection pool size: %s
rica40325-9-15-sftdpo-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-9-15-sftdpo-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-9-15-sftdpo-v4
rica40325-9-15-sftdpo-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-9-15-sftdpo-v4/config.json
rica40325-9-15-sftdpo-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-9-15-sftdpo-v4/special_tokens_map.json
rica40325-9-15-sftdpo-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-9-15-sftdpo-v4/tokenizer_config.json
rica40325-9-15-sftdpo-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-9-15-sftdpo-v4/tokenizer.json
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', '')
Job rica40325-9-15-sftdpo-v4-mkmlizer completed after 79.46s with status: succeeded
Stopping job with name rica40325-9-15-sftdpo-v4-mkmlizer
Pipeline stage MKMLizer completed in 90.70s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 8.12s
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
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rica40325-9-15-sftdpo-v4
Waiting for inference service rica40325-9-15-sftdpo-v4 to be ready
Connection pool is full, discarding connection: %s. Connection pool size: %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', '')
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', '')
Inference service rica40325-9-15-sftdpo-v4 ready after 170.76018595695496s
Pipeline stage MKMLDeployer completed in 173.14s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.7165899276733398s
Received healthy response to inference request in 1.404043436050415s
Received healthy response to inference request in 1.537870168685913s
Received healthy response to inference request in 1.7320001125335693s
Received healthy response to inference request in 2.2121376991271973s
5 requests
0 failed requests
5th percentile: 1.4308087825775146
10th percentile: 1.4575741291046143
20th percentile: 1.5111048221588135
30th percentile: 1.5736141204833984
40th percentile: 1.6451020240783691
50th percentile: 1.7165899276733398
60th percentile: 1.7227540016174316
70th percentile: 1.7289180755615234
80th percentile: 1.828027629852295
90th percentile: 2.0200826644897463
95th percentile: 2.1161101818084718
99th percentile: 2.192932195663452
mean time: 1.7205282688140868
Pipeline stage StressChecker completed in 12.42s
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 4.21s
Shutdown handler de-registered
rica40325-9-15-sftdpo_v4 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 0.14s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-9-15-sftdpo-v4-profiler
Waiting for inference service rica40325-9-15-sftdpo-v4-profiler to be ready
Inference service rica40325-9-15-sftdpo-v4-profiler ready after 170.40607523918152s
Pipeline stage MKMLProfilerDeployer completed in 170.79s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-9-15-sftdpo-v4-profiler-predictor-00001-deployme7ggd2:/code/chaiverse_profiler_1726508656 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-9-15-sftdpo-v4-profiler-predictor-00001-deployme7ggd2 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726508656 && 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_1726508656/summary.json'
kubectl exec -it rica40325-9-15-sftdpo-v4-profiler-predictor-00001-deployme7ggd2 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726508656/summary.json'
Pipeline stage MKMLProfilerRunner completed in 485.45s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-9-15-sftdpo-v4-profiler is running
Tearing down inference service rica40325-9-15-sftdpo-v4-profiler
Service rica40325-9-15-sftdpo-v4-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 9.60s
Shutdown handler de-registered
rica40325-9-15-sftdpo_v4 status is now inactive due to auto deactivation removed underperforming models
rica40325-9-15-sftdpo_v4 status is now torndown due to DeploymentManager action