rica40325-alignment-first

developer_uid: rica40325

submission_id: rica40325-alignment-first_v1

model_name: rica40325-alignment-first_v1

model_group: rica40325/alignment-firs

status: torndown

timestamp: 2024-09-08T08:24:56+00:00

num_battles: 12764

num_wins: 6635

celo_rating: 1255.67

family_friendly_score: 0.0

submission_type: basic

model_repo: rica40325/alignment-first

model_architecture: LlamaForCausalLM

model_num_parameters: 8030261248.0

best_of: 16

max_input_tokens: 512

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.8966162477290761, 'latency_mean': 1.1152388060092926, 'latency_p50': 1.1175297498703003, 'latency_p90': 1.2477927446365356}, {'batch_size': 4, 'throughput': 1.7758985686018414, 'latency_mean': 2.2438090443611145, 'latency_p50': 2.242917060852051, 'latency_p90': 2.5320611953735352}, {'batch_size': 5, 'throughput': 1.8606631219950822, 'latency_mean': 2.675899387598038, 'latency_p50': 2.678977608680725, 'latency_p90': 2.972764730453491}, {'batch_size': 8, 'throughput': 1.9859454965844094, 'latency_mean': 3.992241624593735, 'latency_p50': 3.9761730432510376, 'latency_p90': 4.470172643661499}, {'batch_size': 10, 'throughput': 2.0106809804887598, 'latency_mean': 4.934254194498062, 'latency_p50': 4.937798500061035, 'latency_p90': 5.849210906028747}, {'batch_size': 12, 'throughput': 2.007332689395072, 'latency_mean': 5.890264118909836, 'latency_p50': 5.958246946334839, 'latency_p90': 6.610704612731934}, {'batch_size': 15, 'throughput': 2.007107492811574, 'latency_mean': 7.316469023227691, 'latency_p50': 7.341566443443298, 'latency_p90': 8.189945220947266}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: rica40325-alignment-first_v1

is_internal_developer: False

language_model: rica40325/alignment-first

model_size: 8B

ranking_group: single

throughput_3p7s: 1.98

us_pacific_date: 2024-09-08

win_ratio: 0.5198213726104669

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': 16, '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-alignment-first-v1-mkmlizer
Waiting for job on rica40325-alignment-first-v1-mkmlizer to finish
rica40325-alignment-first-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-alignment-first-v1-mkmlizer: ║     _____            __           __                                ║
rica40325-alignment-first-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rica40325-alignment-first-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rica40325-alignment-first-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rica40325-alignment-first-v1-mkmlizer: ║       /___/                                                         ║
rica40325-alignment-first-v1-mkmlizer: ║                                                                     ║
rica40325-alignment-first-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
rica40325-alignment-first-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rica40325-alignment-first-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rica40325-alignment-first-v1-mkmlizer: ║                                                                     ║
rica40325-alignment-first-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rica40325-alignment-first-v1-mkmlizer: ║  belonging to:                                                      ║
rica40325-alignment-first-v1-mkmlizer: ║                                                                     ║
rica40325-alignment-first-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rica40325-alignment-first-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rica40325-alignment-first-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rica40325-alignment-first-v1-mkmlizer: ║                                                                     ║
rica40325-alignment-first-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Connection pool is full, discarding connection: %s. Connection pool size: %s
rica40325-alignment-first-v1-mkmlizer: Downloaded to shared memory in 37.095s
rica40325-alignment-first-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpxodgbwl6, device:0
rica40325-alignment-first-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Failed to get response for submission chaiml-elo-alignment-run-3_v36: ('http://chaiml-elo-alignment-run-3-v36-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:42996->127.0.0.1:8080: read: connection reset by peer\n')
rica40325-alignment-first-v1-mkmlizer: quantized model in 26.452s
rica40325-alignment-first-v1-mkmlizer: Processed model rica40325/alignment-first in 63.547s
rica40325-alignment-first-v1-mkmlizer: creating bucket guanaco-mkml-models
rica40325-alignment-first-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-alignment-first-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-alignment-first-v1
rica40325-alignment-first-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-alignment-first-v1/config.json
rica40325-alignment-first-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-alignment-first-v1/special_tokens_map.json
rica40325-alignment-first-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-alignment-first-v1/tokenizer_config.json
rica40325-alignment-first-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-alignment-first-v1/tokenizer.json
rica40325-alignment-first-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-alignment-first-v1/flywheel_model.0.safetensors
rica40325-alignment-first-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 49.38it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:04, 60.32it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:03, 67.64it/s]
Loading 0:  12%|█▏        | 34/291 [00:00<00:03, 74.45it/s]
Loading 0:  15%|█▌        | 44/291 [00:00<00:02, 82.59it/s]
Loading 0:  19%|█▉        | 55/291 [00:00<00:02, 91.25it/s]
Loading 0:  22%|██▏       | 65/291 [00:00<00:02, 92.23it/s]
Loading 0:  26%|██▌       | 75/291 [00:00<00:02, 91.68it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:09, 21.93it/s]
Loading 0:  32%|███▏      | 94/291 [00:02<00:07, 27.05it/s]
Loading 0:  35%|███▌      | 103/291 [00:02<00:05, 33.55it/s]
Loading 0:  38%|███▊      | 112/291 [00:02<00:04, 40.35it/s]
Loading 0:  42%|████▏     | 121/291 [00:02<00:03, 47.91it/s]
Loading 0:  45%|████▍     | 130/291 [00:02<00:03, 53.16it/s]
Loading 0:  48%|████▊     | 139/291 [00:02<00:02, 56.95it/s]
Loading 0:  51%|█████     | 148/291 [00:02<00:02, 63.03it/s]
Loading 0:  54%|█████▍    | 157/291 [00:03<00:01, 67.48it/s]
Loading 0:  57%|█████▋    | 166/291 [00:03<00:01, 71.82it/s]
Loading 0:  61%|██████    | 177/291 [00:03<00:01, 81.16it/s]
Loading 0:  64%|██████▍   | 186/291 [00:03<00:01, 78.68it/s]
Loading 0:  67%|██████▋   | 195/291 [00:04<00:04, 21.38it/s]
Loading 0:  69%|██████▉   | 202/291 [00:04<00:03, 25.62it/s]
Loading 0:  73%|███████▎  | 211/291 [00:04<00:02, 32.48it/s]
Loading 0:  76%|███████▌  | 220/291 [00:04<00:01, 38.25it/s]
Loading 0:  79%|███████▊  | 229/291 [00:05<00:01, 45.81it/s]
Loading 0:  82%|████████▏ | 238/291 [00:05<00:01, 52.94it/s]
Loading 0:  85%|████████▍ | 247/291 [00:05<00:00, 58.85it/s]
Loading 0:  88%|████████▊ | 256/291 [00:05<00:00, 64.93it/s]
Loading 0:  91%|█████████ | 265/291 [00:05<00:00, 69.05it/s]
Loading 0:  94%|█████████▍| 274/291 [00:05<00:00, 70.50it/s]
Loading 0:  97%|█████████▋| 283/291 [00:05<00:00, 70.36it/s]
Loading 0: 100%|██████████| 291/291 [00:11<00:00,  5.14it/s]
                                                            
Job rica40325-alignment-first-v1-mkmlizer completed after 84.83s with status: succeeded
Stopping job with name rica40325-alignment-first-v1-mkmlizer
Pipeline stage MKMLizer completed in 86.08s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rica40325-alignment-first-v1
Waiting for inference service rica40325-alignment-first-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
Inference service rica40325-alignment-first-v1 ready after 151.27371311187744s
Pipeline stage MKMLDeployer completed in 151.70s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.5464677810668945s
Received healthy response to inference request in 1.701796531677246s
Received healthy response to inference request in 1.5627241134643555s
Received healthy response to inference request in 1.3760697841644287s
Received healthy response to inference request in 2.0626814365386963s
5 requests
0 failed requests
5th percentile: 1.413400650024414
10th percentile: 1.4507315158843994
20th percentile: 1.5253932476043701
30th percentile: 1.5905385971069337
40th percentile: 1.6461675643920899
50th percentile: 1.701796531677246
60th percentile: 1.8461504936218263
70th percentile: 1.9905044555664062
80th percentile: 2.359438705444336
90th percentile: 2.9529532432556156
95th percentile: 3.2497105121612546
99th percentile: 3.4871163272857664
mean time: 2.0499479293823244
Pipeline stage StressChecker completed in 11.81s
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 4.80s
Shutdown handler de-registered
rica40325-alignment-first_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 0.38s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-alignment-first-v1-profiler
Waiting for inference service rica40325-alignment-first-v1-profiler to be ready
Inference service rica40325-alignment-first-v1-profiler ready after 150.3516652584076s
Pipeline stage MKMLProfilerDeployer completed in 150.72s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-alignment-4d4cda5f7919c8b4e7cae4e1fc5e180c-deploqvjn9:/code/chaiverse_profiler_1725784352 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-alignment-4d4cda5f7919c8b4e7cae4e1fc5e180c-deploqvjn9 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725784352 && 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_1725784352/summary.json'
kubectl exec -it rica40325-alignment-4d4cda5f7919c8b4e7cae4e1fc5e180c-deploqvjn9 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725784352/summary.json'
Pipeline stage MKMLProfilerRunner completed in 851.24s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-alignment-first-v1-profiler is running
Tearing down inference service rica40325-alignment-first-v1-profiler
Service rica40325-alignment-first-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.67s
Shutdown handler de-registered
rica40325-alignment-first_v1 status is now inactive due to auto deactivation removed underperforming models
rica40325-alignment-first_v1 status is now torndown due to DeploymentManager action