stark2000s-dpo-fd-mistral

developer_uid: stark2000s

submission_id: stark2000s-dpo-fd-mistral_v1

model_name: stark2000s-dpo-fd-mistral_v1

model_group: stark2000s/dpo_fd_mistra

status: inactive

timestamp: 2024-12-05T15:27:49+00:00

num_battles: 13661

num_wins: 5181

celo_rating: 1178.24

family_friendly_score: 0.575

family_friendly_standard_error: 0.006991065727054781

submission_type: basic

model_repo: stark2000s/dpo_fd_mistral

model_architecture: LlamaForCausalLM

model_num_parameters: 8030261248.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.8754834849699529, 'latency_mean': 1.1421383011341095, 'latency_p50': 1.1486700773239136, 'latency_p90': 1.2627926111221313}, {'batch_size': 4, 'throughput': 1.8799142381100833, 'latency_mean': 2.1158485317230227, 'latency_p50': 2.1070361137390137, 'latency_p90': 2.3757105350494383}, {'batch_size': 5, 'throughput': 2.047786342292177, 'latency_mean': 2.4297144317626955, 'latency_p50': 2.4163061380386353, 'latency_p90': 2.7443978548049928}, {'batch_size': 8, 'throughput': 2.2514338457727794, 'latency_mean': 3.52493132352829, 'latency_p50': 3.5534058809280396, 'latency_p90': 3.86672899723053}, {'batch_size': 10, 'throughput': 2.3293998698746257, 'latency_mean': 4.255172690153122, 'latency_p50': 4.274770379066467, 'latency_p90': 4.819348955154419}, {'batch_size': 12, 'throughput': 2.3700954510208003, 'latency_mean': 5.007225279808044, 'latency_p50': 5.0128114223480225, 'latency_p90': 5.720911765098572}, {'batch_size': 15, 'throughput': 2.395857286064073, 'latency_mean': 6.1790902531147, 'latency_p50': 6.219233989715576, 'latency_p90': 6.999378561973572}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: stark2000s-dpo-fd-mistral_v1

is_internal_developer: False

language_model: stark2000s/dpo_fd_mistral

model_size: 8B

ranking_group: single

throughput_3p7s: 2.29

us_pacific_date: 2024-12-05

win_ratio: 0.379254812971232

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-dpo-fd-mistral-v1-mkmlizer
Waiting for job on stark2000s-dpo-fd-mistral-v1-mkmlizer to finish
stark2000s-dpo-fd-mistral-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║     _____            __           __                                ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║       /___/                                                         ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║                                                                     ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║  Version: 0.11.12                                                   ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║  https://mk1.ai                                                     ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║                                                                     ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║  belonging to:                                                      ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║                                                                     ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║  Chai Research Corp.                                                ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║  Expiration: 2025-01-15 23:59:59                                    ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ║                                                                     ║
stark2000s-dpo-fd-mistral-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Connection pool is full, discarding connection: %s. Connection pool size: %s
stark2000s-dpo-fd-mistral-v1-mkmlizer: Downloaded to shared memory in 29.379s
stark2000s-dpo-fd-mistral-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpuwmq9kxa, device:0
stark2000s-dpo-fd-mistral-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
stark2000s-dpo-fd-mistral-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-dpo-fd-mistral-v1-mkmlizer:   tensors = torch.load(model_shard_filename, map_location=torch.device(self.device), mmap=True)
stark2000s-dpo-fd-mistral-v1-mkmlizer: quantized model in 25.641s
stark2000s-dpo-fd-mistral-v1-mkmlizer: Processed model stark2000s/dpo_fd_mistral in 55.020s
stark2000s-dpo-fd-mistral-v1-mkmlizer: creating bucket guanaco-mkml-models
stark2000s-dpo-fd-mistral-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
stark2000s-dpo-fd-mistral-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/stark2000s-dpo-fd-mistral-v1
stark2000s-dpo-fd-mistral-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/stark2000s-dpo-fd-mistral-v1/config.json
stark2000s-dpo-fd-mistral-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/stark2000s-dpo-fd-mistral-v1/flywheel_model.0.safetensors
stark2000s-dpo-fd-mistral-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 50.38it/s]
Loading 0:   8%|▊         | 22/291 [00:00<00:03, 85.73it/s]
Loading 0:  11%|█         | 32/291 [00:00<00:02, 90.73it/s]
Loading 0:  14%|█▍        | 42/291 [00:00<00:02, 92.35it/s]
Loading 0:  18%|█▊        | 52/291 [00:00<00:02, 84.56it/s]
Loading 0:  23%|██▎       | 67/291 [00:00<00:02, 92.63it/s]
Loading 0:  26%|██▋       | 77/291 [00:00<00:02, 93.27it/s]
Loading 0:  30%|██▉       | 87/291 [00:01<00:08, 24.55it/s]
Loading 0:  33%|███▎      | 96/291 [00:02<00:06, 30.63it/s]
Loading 0:  36%|███▋      | 106/291 [00:02<00:05, 36.84it/s]
Loading 0:  40%|███▉      | 115/291 [00:02<00:04, 43.83it/s]
Loading 0:  43%|████▎     | 126/291 [00:02<00:03, 54.61it/s]
Loading 0:  48%|████▊     | 139/291 [00:02<00:02, 62.24it/s]
Loading 0:  51%|█████     | 149/291 [00:02<00:02, 69.30it/s]
Loading 0:  54%|█████▍    | 158/291 [00:02<00:01, 73.22it/s]
Loading 0:  58%|█████▊    | 169/291 [00:02<00:01, 74.23it/s]
Loading 0:  63%|██████▎   | 184/291 [00:03<00:01, 84.10it/s]
Loading 0:  67%|██████▋   | 194/291 [00:04<00:03, 26.40it/s]
Loading 0:  69%|██████▉   | 202/291 [00:04<00:02, 31.28it/s]
Loading 0:  74%|███████▎  | 214/291 [00:04<00:01, 39.41it/s]
Loading 0:  77%|███████▋  | 224/291 [00:04<00:01, 47.58it/s]
Loading 0:  80%|████████  | 233/291 [00:04<00:01, 54.43it/s]
Loading 0:  83%|████████▎ | 242/291 [00:04<00:00, 60.17it/s]
Loading 0:  88%|████████▊ | 256/291 [00:04<00:00, 70.54it/s]
Loading 0:  92%|█████████▏| 268/291 [00:05<00:00, 74.76it/s]
Loading 0:  95%|█████████▌| 277/291 [00:05<00:00, 77.29it/s]
Loading 0:  98%|█████████▊| 286/291 [00:05<00:00, 79.17it/s]
                                                            
Job stark2000s-dpo-fd-mistral-v1-mkmlizer completed after 78.31s with status: succeeded
Stopping job with name stark2000s-dpo-fd-mistral-v1-mkmlizer
Pipeline stage MKMLizer completed in 79.54s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.17s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service stark2000s-dpo-fd-mistral-v1
Waiting for inference service stark2000s-dpo-fd-mistral-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
Inference service stark2000s-dpo-fd-mistral-v1 ready after 164.029221534729s
Pipeline stage MKMLDeployer completed in 165.36s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.8424277305603027s
Received healthy response to inference request in 1.7063369750976562s
Received healthy response to inference request in 1.3478822708129883s
Received healthy response to inference request in 1.242272138595581s
Received healthy response to inference request in 1.4092481136322021s
5 requests
0 failed requests
5th percentile: 1.2633941650390625
10th percentile: 1.284516191482544
20th percentile: 1.3267602443695068
30th percentile: 1.3601554393768311
40th percentile: 1.3847017765045166
50th percentile: 1.4092481136322021
60th percentile: 1.5280836582183839
70th percentile: 1.6469192028045654
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
80th percentile: 1.7335551261901856
Connection pool is full, discarding connection: %s. Connection pool size: %s
90th percentile: 1.787991428375244
95th percentile: 1.8152095794677734
99th percentile: 1.8369841003417968
mean time: 1.509633445739746
Pipeline stage StressChecker completed in 9.04s
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 2.47s
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 2.33s
Shutdown handler de-registered
stark2000s-dpo-fd-mistral_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 0.10s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service stark2000s-dpo-fd-mistral-v1-profiler
Waiting for inference service stark2000s-dpo-fd-mistral-v1-profiler to be ready
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 2218.46s
Shutdown handler de-registered
stark2000s-dpo-fd-mistral_v1 status is now inactive due to auto deactivation removed underperforming models