stark2000s-axolotl-ultra_3888

developer_uid: stark2000s

submission_id: stark2000s-axolotl-ultra_3888_v2

model_name: stark2000s-axolotl-ultra_3888_v2

model_group: stark2000s/axolotl-ultra

status: inactive

timestamp: 2024-11-29T01:34:07+00:00

num_battles: 11806

num_wins: 4538

celo_rating: 1176.98

family_friendly_score: 0.5744

family_friendly_standard_error: 0.006992347817435857

submission_type: basic

model_repo: stark2000s/axolotl-ultrachat-default

model_architecture: LlamaForCausalLM

model_num_parameters: 8030261248.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.8725175066872974, 'latency_mean': 1.1460380232334137, 'latency_p50': 1.138296365737915, 'latency_p90': 1.2736486434936523}, {'batch_size': 4, 'throughput': 1.8835851494943676, 'latency_mean': 2.118691169023514, 'latency_p50': 2.1257457733154297, 'latency_p90': 2.381733274459839}, {'batch_size': 5, 'throughput': 2.036332135043099, 'latency_mean': 2.4461194598674774, 'latency_p50': 2.457514762878418, 'latency_p90': 2.7096584320068358}, {'batch_size': 8, 'throughput': 2.2571087064528688, 'latency_mean': 3.5173801589012146, 'latency_p50': 3.517207145690918, 'latency_p90': 3.975407910346985}, {'batch_size': 10, 'throughput': 2.3049192096308064, 'latency_mean': 4.301993200778961, 'latency_p50': 4.326911568641663, 'latency_p90': 4.861863112449646}, {'batch_size': 12, 'throughput': 2.350950006989001, 'latency_mean': 5.063115581274032, 'latency_p50': 5.002102375030518, 'latency_p90': 5.714252352714539}, {'batch_size': 15, 'throughput': 2.366544912693447, 'latency_mean': 6.249746305942535, 'latency_p50': 6.225370287895203, 'latency_p90': 6.966381859779357}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: stark2000s-axolotl-ultra_3888_v2

is_internal_developer: False

language_model: stark2000s/axolotl-ultrachat-default

model_size: 8B

ranking_group: single

throughput_3p7s: 2.29

us_pacific_date: 2024-11-28

win_ratio: 0.38438082331018125

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-axolotl-ultra-3888-v2-mkmlizer
Waiting for job on stark2000s-axolotl-ultra-3888-v2-mkmlizer to finish
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║     _____            __           __                                ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║       /___/                                                         ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║                                                                     ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║  Version: 0.11.12                                                   ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║  https://mk1.ai                                                     ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║                                                                     ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║  belonging to:                                                      ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║                                                                     ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║  Chai Research Corp.                                                ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║  Expiration: 2025-01-15 23:59:59                                    ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ║                                                                     ║
stark2000s-axolotl-ultra-3888-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
stark2000s-axolotl-ultra-3888-v2-mkmlizer: Downloaded to shared memory in 41.360s
stark2000s-axolotl-ultra-3888-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp21esr_ct, device:0
stark2000s-axolotl-ultra-3888-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
stark2000s-axolotl-ultra-3888-v2-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-axolotl-ultra-3888-v2-mkmlizer:   tensors = torch.load(model_shard_filename, map_location=torch.device(self.device), mmap=True)
stark2000s-axolotl-ultra-3888-v2-mkmlizer: quantized model in 27.275s
stark2000s-axolotl-ultra-3888-v2-mkmlizer: Processed model stark2000s/axolotl-ultrachat-default in 68.635s
stark2000s-axolotl-ultra-3888-v2-mkmlizer: creating bucket guanaco-mkml-models
stark2000s-axolotl-ultra-3888-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
stark2000s-axolotl-ultra-3888-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/stark2000s-axolotl-ultra-3888-v2
stark2000s-axolotl-ultra-3888-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/stark2000s-axolotl-ultra-3888-v2/config.json
stark2000s-axolotl-ultra-3888-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/stark2000s-axolotl-ultra-3888-v2/special_tokens_map.json
stark2000s-axolotl-ultra-3888-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/stark2000s-axolotl-ultra-3888-v2/tokenizer_config.json
stark2000s-axolotl-ultra-3888-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/stark2000s-axolotl-ultra-3888-v2/flywheel_model.0.safetensors
stark2000s-axolotl-ultra-3888-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 47.71it/s]
Loading 0:   8%|▊         | 22/291 [00:00<00:03, 79.09it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:03, 78.72it/s]
Loading 0:  14%|█▎        | 40/291 [00:00<00:03, 81.18it/s]
Loading 0:  17%|█▋        | 49/291 [00:00<00:02, 81.64it/s]
Loading 0:  20%|█▉        | 58/291 [00:00<00:02, 80.01it/s]
Loading 0:  23%|██▎       | 67/291 [00:00<00:02, 81.18it/s]
Loading 0:  26%|██▌       | 76/291 [00:00<00:02, 83.04it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:10, 20.59it/s]
Loading 0:  32%|███▏      | 94/291 [00:02<00:07, 26.69it/s]
Loading 0:  35%|███▌      | 103/291 [00:02<00:05, 33.88it/s]
Loading 0:  38%|███▊      | 112/291 [00:02<00:04, 41.72it/s]
Loading 0:  42%|████▏     | 121/291 [00:02<00:03, 47.49it/s]
Loading 0:  45%|████▍     | 130/291 [00:02<00:02, 53.67it/s]
Loading 0:  48%|████▊     | 139/291 [00:02<00:02, 60.78it/s]
Loading 0:  51%|█████     | 148/291 [00:02<00:02, 67.22it/s]
Loading 0:  54%|█████▍    | 158/291 [00:02<00:01, 75.00it/s]
Loading 0:  57%|█████▋    | 167/291 [00:03<00:01, 76.11it/s]
Loading 0:  60%|██████    | 176/291 [00:03<00:01, 76.38it/s]
Loading 0:  64%|██████▎   | 185/291 [00:03<00:01, 78.14it/s]
Loading 0:  67%|██████▋   | 194/291 [00:04<00:04, 20.02it/s]
Loading 0:  69%|██████▉   | 202/291 [00:04<00:03, 24.87it/s]
Loading 0:  73%|███████▎  | 211/291 [00:04<00:02, 30.66it/s]
Loading 0:  76%|███████▋  | 222/291 [00:04<00:01, 40.94it/s]
Loading 0:  80%|███████▉  | 232/291 [00:05<00:01, 46.71it/s]
Loading 0:  83%|████████▎ | 241/291 [00:05<00:00, 52.99it/s]
Loading 0:  86%|████████▌ | 250/291 [00:05<00:00, 58.72it/s]
Loading 0:  90%|████████▉ | 261/291 [00:05<00:00, 69.62it/s]
Loading 0:  94%|█████████▍| 274/291 [00:05<00:00, 76.32it/s]
Loading 0:  98%|█████████▊| 284/291 [00:05<00:00, 81.72it/s]
                                                            
Job stark2000s-axolotl-ultra-3888-v2-mkmlizer completed after 93.57s with status: succeeded
Stopping job with name stark2000s-axolotl-ultra-3888-v2-mkmlizer
Pipeline stage MKMLizer completed in 94.05s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.16s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service stark2000s-axolotl-ultra-3888-v2
Waiting for inference service stark2000s-axolotl-ultra-3888-v2 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-axolotl-ultra-3888-v2 ready after 140.49154257774353s
Pipeline stage MKMLDeployer completed in 141.04s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.69887113571167s
Received healthy response to inference request in 1.4520490169525146s
Received healthy response to inference request in 1.5276777744293213s
Received healthy response to inference request in 1.2594776153564453s
Received healthy response to inference request in 1.7778363227844238s
5 requests
0 failed requests
5th percentile: 1.297991895675659
10th percentile: 1.336506175994873
20th percentile: 1.4135347366333009
30th percentile: 1.467174768447876
40th percentile: 1.4974262714385986
50th percentile: 1.5276777744293213
60th percentile: 1.5961551189422607
70th percentile: 1.6646324634552
80th percentile: 1.7146641731262207
90th percentile: 1.7462502479553224
95th percentile: 1.762043285369873
99th percentile: 1.7746777153015136
mean time: 1.543182373046875
Pipeline stage StressChecker completed in 9.09s
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 2.26s
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 2.03s
Shutdown handler de-registered
stark2000s-axolotl-ultra_3888_v2 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 2627.77s
Shutdown handler de-registered
stark2000s-axolotl-ultra_3888_v2 status is now inactive due to auto deactivation removed underperforming models