jondurbin-airoboros-13b

developer_uid: huohuo12

submission_id: jondurbin-airoboros-13b_v1

model_name: jondurbin-airoboros-13b_v1

model_group: jondurbin/airoboros-13b

status: torndown

timestamp: 2025-02-20T03:07:21+00:00

num_battles: 5792

num_wins: 2197

celo_rating: 1180.61

family_friendly_score: 0.611

family_friendly_standard_error: 0.006894621091836737

submission_type: basic

model_repo: jondurbin/airoboros-13b

model_architecture: LlamaForCausalLM

model_num_parameters: 13015864320.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.4889250369553881, 'latency_mean': 2.0452172136306763, 'latency_p50': 2.030726909637451, 'latency_p90': 2.268285894393921}, {'batch_size': 2, 'throughput': 0.6755683020646213, 'latency_mean': 2.9542419695854187, 'latency_p50': 2.9399741888046265, 'latency_p90': 3.255020833015442}, {'batch_size': 3, 'throughput': 0.78037164478061, 'latency_mean': 3.833407369852066, 'latency_p50': 3.8745301961898804, 'latency_p90': 4.290646719932556}, {'batch_size': 4, 'throughput': 0.8424025551828832, 'latency_mean': 4.725719140768051, 'latency_p50': 4.730284214019775, 'latency_p90': 5.274863219261169}, {'batch_size': 5, 'throughput': 0.8899864064217452, 'latency_mean': 5.594983448982239, 'latency_p50': 5.566394329071045, 'latency_p90': 6.196960687637329}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: jondurbin-airoboros-13b_v1

is_internal_developer: False

language_model: jondurbin/airoboros-13b

model_size: 13B

ranking_group: single

throughput_3p7s: 0.77

us_pacific_date: 2025-02-19

win_ratio: 0.37931629834254144

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 jondurbin-airoboros-13b-v1-mkmlizer
Waiting for job on jondurbin-airoboros-13b-v1-mkmlizer to finish
Failed to get response for submission chaiml-20250219-c-4epoc_80755_v1: HTTPConnectionPool(host='chaiml-20250219-c-4epoc-80755-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
jondurbin-airoboros-13b-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jondurbin-airoboros-13b-v1-mkmlizer: ║     _____            __           __                                ║
jondurbin-airoboros-13b-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jondurbin-airoboros-13b-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jondurbin-airoboros-13b-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jondurbin-airoboros-13b-v1-mkmlizer: ║       /___/                                                         ║
jondurbin-airoboros-13b-v1-mkmlizer: ║                                                                     ║
jondurbin-airoboros-13b-v1-mkmlizer: ║  Version: 0.12.8                                                    ║
jondurbin-airoboros-13b-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jondurbin-airoboros-13b-v1-mkmlizer: ║  https://mk1.ai                                                     ║
jondurbin-airoboros-13b-v1-mkmlizer: ║                                                                     ║
jondurbin-airoboros-13b-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
jondurbin-airoboros-13b-v1-mkmlizer: ║  belonging to:                                                      ║
jondurbin-airoboros-13b-v1-mkmlizer: ║                                                                     ║
jondurbin-airoboros-13b-v1-mkmlizer: ║  Chai Research Corp.                                                ║
jondurbin-airoboros-13b-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jondurbin-airoboros-13b-v1-mkmlizer: ║  Expiration: 2025-04-15 23:59:59                                    ║
jondurbin-airoboros-13b-v1-mkmlizer: ║                                                                     ║
jondurbin-airoboros-13b-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jondurbin-airoboros-13b-v1-mkmlizer: Downloaded to shared memory in 85.802s
jondurbin-airoboros-13b-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp8v33x03h, device:0
jondurbin-airoboros-13b-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jondurbin-airoboros-13b-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.
jondurbin-airoboros-13b-v1-mkmlizer:   tensors = torch.load(model_shard_filename, map_location=torch.device(self.device), mmap=True)
jondurbin-airoboros-13b-v1-mkmlizer: 
Loading 0:   0%|          | 0/403 [00:00<?, ?it/s]
Loading 0:   1%|          | 4/403 [00:00<00:11, 34.99it/s]
Loading 0:   3%|▎         | 14/403 [00:00<00:07, 52.35it/s]
Loading 0:   6%|▌         | 24/403 [00:00<00:06, 59.02it/s]
Loading 0:   8%|▊         | 34/403 [00:00<00:06, 60.53it/s]
Loading 0:  11%|█         | 44/403 [00:00<00:05, 60.26it/s]
Loading 0:  13%|█▎        | 54/403 [00:00<00:05, 61.14it/s]
Loading 0:  16%|█▌        | 64/403 [00:01<00:05, 61.58it/s]
Loading 0:  18%|█▊        | 74/403 [00:01<00:05, 60.17it/s]
Loading 0:  20%|██        | 81/403 [00:03<00:26, 12.04it/s]
Loading 0:  22%|██▏       | 89/403 [00:03<00:20, 15.33it/s]
Loading 0:  25%|██▍       | 99/403 [00:03<00:14, 20.37it/s]
Loading 0:  27%|██▋       | 109/403 [00:03<00:11, 26.20it/s]
Loading 0:  30%|██▉       | 119/403 [00:03<00:08, 32.15it/s]
Loading 0:  32%|███▏      | 129/403 [00:04<00:07, 37.22it/s]
Loading 0:  34%|███▍      | 139/403 [00:04<00:06, 42.13it/s]
Loading 0:  37%|███▋      | 149/403 [00:04<00:05, 47.12it/s]
Loading 0:  38%|███▊      | 155/403 [00:06<00:21, 11.79it/s]
Loading 0:  41%|████      | 164/403 [00:06<00:15, 15.57it/s]
Loading 0:  43%|████▎     | 174/403 [00:06<00:11, 20.74it/s]
Loading 0:  46%|████▌     | 184/403 [00:06<00:08, 26.18it/s]
Loading 0:  48%|████▊     | 194/403 [00:07<00:06, 31.83it/s]
Loading 0:  51%|█████     | 204/403 [00:07<00:05, 37.67it/s]
Loading 0:  53%|█████▎    | 214/403 [00:07<00:04, 42.75it/s]
Loading 0:  56%|█████▌    | 224/403 [00:07<00:03, 46.85it/s]
Loading 0:  58%|█████▊    | 232/403 [00:09<00:13, 12.54it/s]
Loading 0:  59%|█████▉    | 237/403 [00:09<00:11, 14.50it/s]
Loading 0:  61%|██████    | 244/403 [00:09<00:08, 17.71it/s]
Loading 0:  63%|██████▎   | 254/403 [00:09<00:06, 23.70it/s]
Loading 0:  66%|██████▌   | 264/403 [00:10<00:04, 29.55it/s]
Loading 0:  68%|██████▊   | 274/403 [00:10<00:03, 34.97it/s]
Loading 0:  70%|███████   | 284/403 [00:10<00:02, 40.34it/s]
Loading 0:  73%|███████▎  | 294/403 [00:10<00:02, 44.52it/s]
Loading 0:  75%|███████▌  | 304/403 [00:10<00:02, 48.68it/s]
Loading 0:  77%|███████▋  | 310/403 [00:12<00:07, 11.76it/s]
Loading 0:  79%|███████▉  | 319/403 [00:12<00:05, 15.56it/s]
Loading 0:  82%|████████▏ | 329/403 [00:13<00:03, 20.51it/s]
Loading 0:  84%|████████▍ | 339/403 [00:13<00:02, 26.12it/s]
Loading 0:  87%|████████▋ | 349/403 [00:13<00:01, 31.92it/s]
Loading 0:  89%|████████▉ | 359/403 [00:13<00:01, 37.10it/s]
Loading 0:  92%|█████████▏| 369/403 [00:13<00:00, 42.36it/s]
Loading 0:  94%|█████████▍| 379/403 [00:13<00:00, 46.86it/s]
Loading 0:  96%|█████████▋| 388/403 [00:14<00:00, 25.17it/s]
Loading 0:  98%|█████████▊| 394/403 [00:14<00:00, 27.11it/s]
Loading 0: 100%|██████████| 403/403 [00:16<00:00, 12.10it/s]
                                                            
You are using the default legacy behaviour of the <class 'transformers.models.llama.tokenization_llama.LlamaTokenizer'>. This is expected, and simply means that the `legacy` (previous) behavior will be used so nothing changes for you. If you want to use the new behaviour, set `legacy=False`. This should only be set if you understand what it means, and thoroughly read the reason why this was added as explained in https://github.com/huggingface/transformers/pull/24565 - if you loaded a llama tokenizer from a GGUF file you can ignore this message
jondurbin-airoboros-13b-v1-mkmlizer: You are using the default legacy behaviour of the <class 'transformers.models.llama.tokenization_llama_fast.LlamaTokenizerFast'>. This is expected, and simply means that the `legacy` (previous) behavior will be used so nothing changes for you. If you want to use the new behaviour, set `legacy=False`. This should only be set if you understand what it means, and thoroughly read the reason why this was added as explained in https://github.com/huggingface/transformers/pull/24565 - if you loaded a llama tokenizer from a GGUF file you can ignore this message.
jondurbin-airoboros-13b-v1-mkmlizer: quantized model in 35.514s
jondurbin-airoboros-13b-v1-mkmlizer: Processed model jondurbin/airoboros-13b in 121.318s
jondurbin-airoboros-13b-v1-mkmlizer: creating bucket guanaco-mkml-models
jondurbin-airoboros-13b-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jondurbin-airoboros-13b-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jondurbin-airoboros-13b-v1
jondurbin-airoboros-13b-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jondurbin-airoboros-13b-v1/config.json
jondurbin-airoboros-13b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jondurbin-airoboros-13b-v1/tokenizer_config.json
jondurbin-airoboros-13b-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jondurbin-airoboros-13b-v1/special_tokens_map.json
jondurbin-airoboros-13b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/jondurbin-airoboros-13b-v1/tokenizer.model
jondurbin-airoboros-13b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jondurbin-airoboros-13b-v1/tokenizer.json
jondurbin-airoboros-13b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jondurbin-airoboros-13b-v1/flywheel_model.0.safetensors
Job jondurbin-airoboros-13b-v1-mkmlizer completed after 145.31s with status: succeeded
Stopping job with name jondurbin-airoboros-13b-v1-mkmlizer
Pipeline stage MKMLizer completed in 145.83s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.14s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service jondurbin-airoboros-13b-v1
Waiting for inference service jondurbin-airoboros-13b-v1 to be ready
Inference service jondurbin-airoboros-13b-v1 ready after 180.85480666160583s
Pipeline stage MKMLDeployer completed in 181.37s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.4520394802093506s
Received healthy response to inference request in 1.2394688129425049s
Received healthy response to inference request in 1.669895887374878s
Received healthy response to inference request in 1.7071712017059326s
Received healthy response to inference request in 1.7238671779632568s
5 requests
0 failed requests
5th percentile: 1.3255542278289796
10th percentile: 1.411639642715454
20th percentile: 1.5838104724884032
30th percentile: 1.6773509502410888
40th percentile: 1.6922610759735108
50th percentile: 1.7071712017059326
60th percentile: 1.7138495922088623
70th percentile: 1.720527982711792
80th percentile: 1.8695016384124756
90th percentile: 2.160770559310913
95th percentile: 2.3064050197601316
99th percentile: 2.4229125881195066
mean time: 1.7584885120391847
Pipeline stage StressChecker completed in 10.09s
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 0.67s
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 0.61s
Shutdown handler de-registered
jondurbin-airoboros-13b_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 0.10s
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 jondurbin-airoboros-13b-v1-profiler
Waiting for inference service jondurbin-airoboros-13b-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
%s, retrying in %s seconds...
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 5305.31s
Shutdown handler de-registered
jondurbin-airoboros-13b_v1 status is now inactive due to auto deactivation removed underperforming models
jondurbin-airoboros-13b_v1 status is now torndown due to DeploymentManager action
jondurbin-airoboros-13b_v1 status is now torndown due to DeploymentManager action
jondurbin-airoboros-13b_v1 status is now torndown due to DeploymentManager action