stark2000s-llama-3-test

developer_uid: stark2000s

submission_id: stark2000s-llama-3-test_v1

model_name: stark2000s-llama-3-test_v1

model_group: stark2000s/llama-3_test

status: inactive

timestamp: 2024-12-12T09:55:51+00:00

num_battles: 10406

num_wins: 4415

celo_rating: 1212.8

family_friendly_score: 0.5664

family_friendly_standard_error: 0.007008438342455472

submission_type: basic

model_repo: stark2000s/llama-3_test

model_architecture: LlamaForCausalLM

model_num_parameters: 8030261248.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.8729886464212281, 'latency_mean': 1.1452260601520539, 'latency_p50': 1.1389329433441162, 'latency_p90': 1.2661732196807862}, {'batch_size': 4, 'throughput': 1.9148765306887856, 'latency_mean': 2.079357233047485, 'latency_p50': 2.084740996360779, 'latency_p90': 2.281309986114502}, {'batch_size': 5, 'throughput': 2.066255218563525, 'latency_mean': 2.414946858882904, 'latency_p50': 2.42866587638855, 'latency_p90': 2.694259524345398}, {'batch_size': 8, 'throughput': 2.316526779034797, 'latency_mean': 3.432683789730072, 'latency_p50': 3.4229044914245605, 'latency_p90': 3.8731159925460816}, {'batch_size': 10, 'throughput': 2.3742992907060056, 'latency_mean': 4.187894994020462, 'latency_p50': 4.159778594970703, 'latency_p90': 4.747934556007385}, {'batch_size': 12, 'throughput': 2.3751385280260204, 'latency_mean': 5.008845398426056, 'latency_p50': 5.020666837692261, 'latency_p90': 5.713057374954223}, {'batch_size': 15, 'throughput': 2.429787876789359, 'latency_mean': 6.089861125946045, 'latency_p50': 6.116901159286499, 'latency_p90': 6.779440093040466}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: stark2000s-llama-3-test_v1

is_internal_developer: False

language_model: stark2000s/llama-3_test

model_size: 8B

ranking_group: single

throughput_3p7s: 2.35

us_pacific_date: 2024-12-12

win_ratio: 0.42427445704401306

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': 1024, '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 stark2000s-llama-3-test-v1-mkmlizer
Waiting for job on stark2000s-llama-3-test-v1-mkmlizer to finish
stark2000s-llama-3-test-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
stark2000s-llama-3-test-v1-mkmlizer: ║     _____            __           __                                ║
stark2000s-llama-3-test-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
stark2000s-llama-3-test-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
stark2000s-llama-3-test-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
stark2000s-llama-3-test-v1-mkmlizer: ║       /___/                                                         ║
stark2000s-llama-3-test-v1-mkmlizer: ║                                                                     ║
stark2000s-llama-3-test-v1-mkmlizer: ║  Version: 0.11.12                                                   ║
stark2000s-llama-3-test-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
stark2000s-llama-3-test-v1-mkmlizer: ║  https://mk1.ai                                                     ║
stark2000s-llama-3-test-v1-mkmlizer: ║                                                                     ║
stark2000s-llama-3-test-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
stark2000s-llama-3-test-v1-mkmlizer: ║  belonging to:                                                      ║
stark2000s-llama-3-test-v1-mkmlizer: ║                                                                     ║
stark2000s-llama-3-test-v1-mkmlizer: ║  Chai Research Corp.                                                ║
stark2000s-llama-3-test-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
stark2000s-llama-3-test-v1-mkmlizer: ║  Expiration: 2025-01-15 23:59:59                                    ║
stark2000s-llama-3-test-v1-mkmlizer: ║                                                                     ║
stark2000s-llama-3-test-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
stark2000s-llama-3-test-v1-mkmlizer: Downloaded to shared memory in 34.245s
stark2000s-llama-3-test-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpmgj7rrt9, device:0
stark2000s-llama-3-test-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
stark2000s-llama-3-test-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/mk1/flywheel/functional/loader.py:55: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
stark2000s-llama-3-test-v1-mkmlizer:   tensors = torch.load(model_shard_filename, map_location=torch.device(self.device), mmap=True)
stark2000s-llama-3-test-v1-mkmlizer: quantized model in 27.742s
stark2000s-llama-3-test-v1-mkmlizer: Processed model stark2000s/llama-3_test in 61.987s
stark2000s-llama-3-test-v1-mkmlizer: creating bucket guanaco-mkml-models
stark2000s-llama-3-test-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
stark2000s-llama-3-test-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/stark2000s-llama-3-test-v1
stark2000s-llama-3-test-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/stark2000s-llama-3-test-v1/special_tokens_map.json
stark2000s-llama-3-test-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/stark2000s-llama-3-test-v1/config.json
stark2000s-llama-3-test-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/stark2000s-llama-3-test-v1/tokenizer_config.json
stark2000s-llama-3-test-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/stark2000s-llama-3-test-v1/tokenizer.json
stark2000s-llama-3-test-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/stark2000s-llama-3-test-v1/flywheel_model.0.safetensors
stark2000s-llama-3-test-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 48.44it/s]
Loading 0:   8%|▊         | 22/291 [00:00<00:03, 81.57it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:03, 83.26it/s]
Loading 0:  14%|█▍        | 42/291 [00:00<00:02, 91.89it/s]
Loading 0:  18%|█▊        | 52/291 [00:00<00:02, 83.33it/s]
Loading 0:  23%|██▎       | 67/291 [00:00<00:02, 91.26it/s]
Loading 0:  26%|██▋       | 77/291 [00:00<00:02, 90.59it/s]
Loading 0:  30%|██▉       | 87/291 [00:02<00:09, 22.35it/s]
Loading 0:  32%|███▏      | 94/291 [00:02<00:07, 25.42it/s]
Loading 0:  35%|███▌      | 103/291 [00:02<00:05, 31.60it/s]
Loading 0:  38%|███▊      | 112/291 [00:02<00:04, 38.72it/s]
Loading 0:  42%|████▏     | 121/291 [00:02<00:03, 44.12it/s]
Loading 0:  45%|████▍     | 130/291 [00:02<00:03, 49.76it/s]
Loading 0:  48%|████▊     | 139/291 [00:02<00:02, 56.56it/s]
Loading 0:  51%|█████     | 148/291 [00:03<00:02, 59.33it/s]
Loading 0:  54%|█████▍    | 157/291 [00:03<00:02, 65.45it/s]
Loading 0:  57%|█████▋    | 166/291 [00:03<00:01, 67.86it/s]
Loading 0:  60%|██████    | 175/291 [00:03<00:01, 70.96it/s]
Loading 0:  63%|██████▎   | 184/291 [00:03<00:01, 75.38it/s]
Loading 0:  66%|██████▋   | 193/291 [00:04<00:05, 19.49it/s]
Loading 0:  69%|██████▉   | 202/291 [00:04<00:03, 25.22it/s]
Loading 0:  73%|███████▎  | 211/291 [00:04<00:02, 31.87it/s]
Loading 0:  76%|███████▌  | 220/291 [00:05<00:01, 39.17it/s]
Loading 0:  79%|███████▉  | 230/291 [00:05<00:01, 48.72it/s]
Loading 0:  83%|████████▎ | 241/291 [00:05<00:00, 54.79it/s]
Loading 0:  86%|████████▌ | 250/291 [00:05<00:00, 61.14it/s]
Loading 0:  89%|████████▉ | 259/291 [00:05<00:00, 64.98it/s]
Loading 0:  92%|█████████▏| 268/291 [00:05<00:00, 70.59it/s]
Loading 0:  96%|█████████▌| 278/291 [00:05<00:00, 77.79it/s]
Loading 0:  99%|█████████▊| 287/291 [00:06<00:00, 40.64it/s]
                                                            
Job stark2000s-llama-3-test-v1-mkmlizer completed after 94.53s with status: succeeded
Stopping job with name stark2000s-llama-3-test-v1-mkmlizer
Pipeline stage MKMLizer completed in 95.03s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.72s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service stark2000s-llama-3-test-v1
Waiting for inference service stark2000s-llama-3-test-v1 to be ready
Failed to get response for submission chaiml-mistral-nemo-dpo-_8035_v1: ('http://chaiml-mistral-nemo-dpo-8035-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:50672->127.0.0.1:8080: read: connection reset by peer\n')
Inference service stark2000s-llama-3-test-v1 ready after 190.7709653377533s
Pipeline stage MKMLDeployer completed in 191.31s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.6022937297821045s
Received healthy response to inference request in 1.4445912837982178s
Received healthy response to inference request in 1.50352144241333s
Received healthy response to inference request in 1.2231171131134033s
Received healthy response to inference request in 1.2292256355285645s
5 requests
0 failed requests
5th percentile: 1.2243388175964356
10th percentile: 1.2255605220794679
20th percentile: 1.2280039310455322
30th percentile: 1.272298765182495
40th percentile: 1.3584450244903565
50th percentile: 1.4445912837982178
60th percentile: 1.4681633472442628
70th percentile: 1.4917354106903076
80th percentile: 1.523275899887085
90th percentile: 1.5627848148345946
95th percentile: 1.5825392723083496
99th percentile: 1.5983428382873535
mean time: 1.400549840927124
Pipeline stage StressChecker completed in 8.32s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyTriggerPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage OfflineFamilyFriendlyTriggerPipeline completed in 2.14s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 2.05s
Shutdown handler de-registered
stark2000s-llama-3-test_v1 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyScorer
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 2230.47s
Shutdown handler de-registered
stark2000s-llama-3-test_v1 status is now inactive due to auto deactivation removed underperforming models