developer_uid: robert_irvine
submission_id: jic062-instruct_v16
model_name: jic062-instruct_v16
model_group: jic062/instruct
status: torndown
timestamp: 2024-08-05T00:48:39+00:00
num_battles: 168434
num_wins: 82879
celo_rating: 1197.52
family_friendly_score: 0.0
submission_type: basic
model_repo: jic062/instruct
model_architecture: LlamaForCausalLM
reward_repo: ChaiML/gpt2_xl_pairwise_89m_step_347634
model_num_parameters: 8030261248.0
best_of: 4
max_input_tokens: 512
max_output_tokens: 64
display_name: jic062-instruct_v16
is_internal_developer: True
language_model: jic062/instruct
model_size: 8B
ranking_group: single
us_pacific_date: 2024-08-04
win_ratio: 0.49205623567688234
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', '|eot_id|', '<|end_of_text|>'], 'max_input_tokens': 512, 'best_of': 4, 'max_output_tokens': 64, 'reward_max_token_input': 256}
formatter: {'memory_template': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}:', 'truncate_by_message': False}
reward_formatter: {'bot_template': 'Bot: {message}\n', 'memory_template': '', 'prompt_template': '', 'response_template': 'Bot:', 'truncate_by_message': False, 'user_template': 'User: {message}\n'}
Resubmit model
Running pipeline stage MKMLizer
Starting job with name jic062-instruct-v16-mkmlizer
Waiting for job on jic062-instruct-v16-mkmlizer to finish
jic062-instruct-v16-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-instruct-v16-mkmlizer: ║     _____            __           __                                ║
jic062-instruct-v16-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-instruct-v16-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-instruct-v16-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-instruct-v16-mkmlizer: ║       /___/                                                         ║
jic062-instruct-v16-mkmlizer: ║                                                                     ║
jic062-instruct-v16-mkmlizer: ║  Version: 0.9.9                                                     ║
jic062-instruct-v16-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-instruct-v16-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-instruct-v16-mkmlizer: ║                                                                     ║
jic062-instruct-v16-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-instruct-v16-mkmlizer: ║  belonging to:                                                      ║
jic062-instruct-v16-mkmlizer: ║                                                                     ║
jic062-instruct-v16-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-instruct-v16-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-instruct-v16-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-instruct-v16-mkmlizer: ║                                                                     ║
jic062-instruct-v16-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-instruct-v16-mkmlizer: Downloaded to shared memory in 32.384s
jic062-instruct-v16-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpk_4xni_1, device:0
jic062-instruct-v16-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-instruct-v16-mkmlizer: quantized model in 26.119s
jic062-instruct-v16-mkmlizer: Processed model jic062/instruct in 58.504s
jic062-instruct-v16-mkmlizer: creating bucket guanaco-mkml-models
jic062-instruct-v16-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-instruct-v16-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-instruct-v16
jic062-instruct-v16-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-instruct-v16/config.json
jic062-instruct-v16-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-instruct-v16/special_tokens_map.json
jic062-instruct-v16-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-instruct-v16/tokenizer_config.json
jic062-instruct-v16-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-instruct-v16/tokenizer.json
jic062-instruct-v16-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
jic062-instruct-v16-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 48.43it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:04, 67.14it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:03, 69.25it/s]
Loading 0:  12%|█▏        | 34/291 [00:00<00:03, 75.77it/s]
Loading 0:  17%|█▋        | 49/291 [00:00<00:02, 86.46it/s]
Loading 0:  20%|█▉        | 58/291 [00:00<00:02, 84.63it/s]
Loading 0:  23%|██▎       | 67/291 [00:00<00:02, 81.32it/s]
Loading 0:  26%|██▌       | 76/291 [00:00<00:02, 79.59it/s]
Loading 0:  29%|██▉       | 84/291 [00:02<00:09, 21.49it/s]
Loading 0:  31%|███       | 90/291 [00:02<00:08, 24.79it/s]
Loading 0:  33%|███▎      | 97/291 [00:02<00:06, 30.02it/s]
Loading 0:  36%|███▋      | 106/291 [00:02<00:04, 37.67it/s]
Loading 0:  40%|███▉      | 116/291 [00:02<00:03, 47.93it/s]
Loading 0:  43%|████▎     | 125/291 [00:02<00:02, 55.99it/s]
Loading 0:  48%|████▊     | 139/291 [00:02<00:02, 67.43it/s]
Loading 0:  51%|█████     | 148/291 [00:02<00:02, 68.86it/s]
Loading 0:  54%|█████▍    | 157/291 [00:02<00:01, 72.33it/s]
Loading 0:  58%|█████▊    | 169/291 [00:03<00:01, 74.28it/s]
Loading 0:  61%|██████    | 178/291 [00:03<00:01, 77.48it/s]
Loading 0:  64%|██████▍   | 187/291 [00:04<00:04, 22.83it/s]
Loading 0:  68%|██████▊   | 197/291 [00:04<00:03, 29.84it/s]
Loading 0:  71%|███████   | 206/291 [00:04<00:02, 36.71it/s]
Loading 0:  74%|███████▎  | 214/291 [00:04<00:01, 42.61it/s]
Loading 0:  77%|███████▋  | 223/291 [00:04<00:01, 49.42it/s]
Loading 0:  80%|████████  | 233/291 [00:04<00:00, 58.87it/s]
Loading 0:  83%|████████▎ | 242/291 [00:04<00:00, 65.00it/s]
Loading 0:  86%|████████▋ | 251/291 [00:05<00:00, 68.27it/s]
Loading 0:  89%|████████▉ | 260/291 [00:05<00:00, 71.70it/s]
Loading 0:  92%|█████████▏| 269/291 [00:05<00:00, 76.31it/s]
Loading 0:  96%|█████████▌| 278/291 [00:05<00:00, 79.67it/s]
Loading 0:  99%|█████████▊| 287/291 [00:05<00:00, 42.22it/s]
                                                            
/opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:957: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jic062-instruct-v16-mkmlizer:   warnings.warn(
jic062-instruct-v16-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:785: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jic062-instruct-v16-mkmlizer:   warnings.warn(
jic062-instruct-v16-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:469: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jic062-instruct-v16-mkmlizer:   warnings.warn(
jic062-instruct-v16-mkmlizer: 
Downloading shards:   0%|          | 0/2 [00:00<?, ?it/s]
Downloading shards:  50%|█████     | 1/2 [00:05<00:05,  5.54s/it]
Downloading shards: 100%|██████████| 2/2 [00:08<00:00,  4.05s/it]
Downloading shards: 100%|██████████| 2/2 [00:08<00:00,  4.27s/it]
jic062-instruct-v16-mkmlizer: 
Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]
Loading checkpoint shards:  50%|█████     | 1/2 [00:00<00:00,  2.40it/s]
Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  3.98it/s]
Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  3.62it/s]
jic062-instruct-v16-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jic062-instruct-v16-mkmlizer: Saving duration: 1.342s
jic062-instruct-v16-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 13.388s
jic062-instruct-v16-mkmlizer: creating bucket guanaco-reward-models
jic062-instruct-v16-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jic062-instruct-v16-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jic062-instruct-v16_reward
jic062-instruct-v16-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jic062-instruct-v16_reward/config.json
jic062-instruct-v16-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jic062-instruct-v16_reward/special_tokens_map.json
jic062-instruct-v16-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jic062-instruct-v16_reward/tokenizer_config.json
jic062-instruct-v16-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jic062-instruct-v16_reward/merges.txt
jic062-instruct-v16-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jic062-instruct-v16_reward/vocab.json
jic062-instruct-v16-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jic062-instruct-v16_reward/tokenizer.json
Job jic062-instruct-v16-mkmlizer completed after 105.01s with status: succeeded
Stopping job with name jic062-instruct-v16-mkmlizer
Pipeline stage MKMLizer completed in 106.07s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service jic062-instruct-v16
Waiting for inference service jic062-instruct-v16 to be ready
Inference service jic062-instruct-v16 ready after 150.91198253631592s
Pipeline stage ISVCDeployer completed in 152.63s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.050095319747925s
Received healthy response to inference request in 1.1258623600006104s
Received healthy response to inference request in 1.105501413345337s
Received healthy response to inference request in 1.094646692276001s
Received healthy response to inference request in 1.1151690483093262s
5 requests
0 failed requests
5th percentile: 1.096817636489868
10th percentile: 1.0989885807037354
20th percentile: 1.1033304691314698
30th percentile: 1.1074349403381347
40th percentile: 1.1113019943237306
50th percentile: 1.1151690483093262
60th percentile: 1.11944637298584
70th percentile: 1.1237236976623535
80th percentile: 1.3107089519500734
90th percentile: 1.6804021358489991
95th percentile: 1.8652487277984617
99th percentile: 2.013126001358032
mean time: 1.2982549667358398
Pipeline stage StressChecker completed in 7.45s
jic062-instruct_v16 status is now deployed due to DeploymentManager action
jic062-instruct_v16 status is now inactive due to auto deactivation removed underperforming models
jic062-instruct_v16 status is now torndown due to DeploymentManager action