developer_uid: chace9580
submission_id: jic062-instruct-v17-g4_v2
model_name: jic062-instruct-v17-g4_v2
model_group: jic062/instruct_v17_g4
status: torndown
timestamp: 2024-08-04T03:52:29+00:00
num_battles: 14975
num_wins: 7846
celo_rating: 1228.22
family_friendly_score: 0.0
submission_type: basic
model_repo: jic062/instruct_v17_g4
model_architecture: LlamaForCausalLM
reward_repo: ChaiML/gpt2_xl_pairwise_89m_step_347634
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: jic062-instruct-v17-g4_v2
is_internal_developer: False
language_model: jic062/instruct_v17_g4
model_size: 8B
ranking_group: single
us_pacific_date: 2024-08-03
win_ratio: 0.5239398998330551
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', '<|end_of_text|>', '|eot_id|'], 'max_input_tokens': 512, 'best_of': 16, '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_name}: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '{user_name}: {message}\n'}
Resubmit model
Running pipeline stage MKMLizer
Starting job with name jic062-instruct-v17-g4-v2-mkmlizer
Waiting for job on jic062-instruct-v17-g4-v2-mkmlizer to finish
jic062-instruct-v17-g4-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-instruct-v17-g4-v2-mkmlizer: ║     _____            __           __                                ║
jic062-instruct-v17-g4-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-instruct-v17-g4-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-instruct-v17-g4-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-instruct-v17-g4-v2-mkmlizer: ║       /___/                                                         ║
jic062-instruct-v17-g4-v2-mkmlizer: ║                                                                     ║
jic062-instruct-v17-g4-v2-mkmlizer: ║  Version: 0.9.9                                                     ║
jic062-instruct-v17-g4-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-instruct-v17-g4-v2-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-instruct-v17-g4-v2-mkmlizer: ║                                                                     ║
jic062-instruct-v17-g4-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-instruct-v17-g4-v2-mkmlizer: ║  belonging to:                                                      ║
jic062-instruct-v17-g4-v2-mkmlizer: ║                                                                     ║
jic062-instruct-v17-g4-v2-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-instruct-v17-g4-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-instruct-v17-g4-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-instruct-v17-g4-v2-mkmlizer: ║                                                                     ║
jic062-instruct-v17-g4-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-instruct-v17-g4-v2-mkmlizer: Downloaded to shared memory in 22.188s
jic062-instruct-v17-g4-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmplgmho_3t, device:0
jic062-instruct-v17-g4-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-instruct-v17-g4-v2-mkmlizer: quantized model in 26.658s
jic062-instruct-v17-g4-v2-mkmlizer: Processed model jic062/instruct_v17_g4 in 48.846s
jic062-instruct-v17-g4-v2-mkmlizer: creating bucket guanaco-mkml-models
jic062-instruct-v17-g4-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-instruct-v17-g4-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-instruct-v17-g4-v2
jic062-instruct-v17-g4-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-instruct-v17-g4-v2/config.json
jic062-instruct-v17-g4-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-instruct-v17-g4-v2/special_tokens_map.json
jic062-instruct-v17-g4-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-instruct-v17-g4-v2/tokenizer_config.json
jic062-instruct-v17-g4-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-instruct-v17-g4-v2/tokenizer.json
jic062-instruct-v17-g4-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-instruct-v17-g4-v2/flywheel_model.0.safetensors
jic062-instruct-v17-g4-v2-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
jic062-instruct-v17-g4-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 51.05it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:04, 67.56it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:03, 75.15it/s]
Loading 0:  12%|█▏        | 34/291 [00:00<00:03, 75.73it/s]
Loading 0:  15%|█▍        | 43/291 [00:00<00:03, 77.07it/s]
Loading 0:  18%|█▊        | 52/291 [00:00<00:03, 77.54it/s]
Loading 0:  21%|██        | 61/291 [00:00<00:02, 78.99it/s]
Loading 0:  24%|██▍       | 70/291 [00:00<00:02, 78.79it/s]
Loading 0:  27%|██▋       | 79/291 [00:01<00:02, 78.73it/s]
Loading 0:  30%|██▉       | 87/291 [00:02<00:09, 20.53it/s]
Loading 0:  32%|███▏      | 94/291 [00:02<00:08, 24.28it/s]
Loading 0:  35%|███▌      | 103/291 [00:02<00:06, 30.92it/s]
Loading 0:  38%|███▊      | 112/291 [00:02<00:04, 38.00it/s]
Loading 0:  42%|████▏     | 121/291 [00:02<00:03, 45.67it/s]
Loading 0:  45%|████▍     | 130/291 [00:02<00:03, 53.23it/s]
Loading 0:  48%|████▊     | 139/291 [00:02<00:02, 58.84it/s]
Loading 0:  51%|█████     | 148/291 [00:02<00:02, 64.56it/s]
Loading 0:  54%|█████▍    | 157/291 [00:03<00:01, 67.78it/s]
Loading 0:  57%|█████▋    | 166/291 [00:03<00:01, 70.55it/s]
Loading 0:  60%|██████    | 175/291 [00:03<00:01, 74.21it/s]
Loading 0:  63%|██████▎   | 184/291 [00:03<00:01, 74.29it/s]
Loading 0:  66%|██████▌   | 192/291 [00:04<00:04, 20.72it/s]
Loading 0:  68%|██████▊   | 198/291 [00:04<00:03, 24.18it/s]
Loading 0:  70%|███████   | 205/291 [00:04<00:02, 29.07it/s]
Loading 0:  74%|███████▎  | 214/291 [00:04<00:02, 36.71it/s]
Loading 0:  77%|███████▋  | 223/291 [00:05<00:01, 43.91it/s]
Loading 0:  80%|███████▉  | 232/291 [00:05<00:01, 50.59it/s]
Loading 0:  83%|████████▎ | 241/291 [00:05<00:00, 56.69it/s]
Loading 0:  86%|████████▌ | 250/291 [00:05<00:00, 61.71it/s]
Loading 0:  89%|████████▉ | 259/291 [00:05<00:00, 66.86it/s]
Loading 0:  92%|█████████▏| 268/291 [00:05<00:00, 69.72it/s]
Loading 0:  95%|█████████▌| 277/291 [00:05<00:00, 72.75it/s]
Loading 0:  98%|█████████▊| 286/291 [00:05<00:00, 73.19it/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-v17-g4-v2-mkmlizer:   warnings.warn(
jic062-instruct-v17-g4-v2-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-v17-g4-v2-mkmlizer:   warnings.warn(
jic062-instruct-v17-g4-v2-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-v17-g4-v2-mkmlizer:   warnings.warn(
jic062-instruct-v17-g4-v2-mkmlizer: 
Downloading shards:   0%|          | 0/2 [00:00<?, ?it/s]
Downloading shards:  50%|█████     | 1/2 [00:05<00:05,  5.33s/it]
Downloading shards: 100%|██████████| 2/2 [00:08<00:00,  4.04s/it]
Downloading shards: 100%|██████████| 2/2 [00:08<00:00,  4.24s/it]
jic062-instruct-v17-g4-v2-mkmlizer: 
Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]
Loading checkpoint shards:  50%|█████     | 1/2 [00:00<00:00,  2.42it/s]
Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  3.95it/s]
Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  3.61it/s]
jic062-instruct-v17-g4-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jic062-instruct-v17-g4-v2-mkmlizer: Saving duration: 1.344s
jic062-instruct-v17-g4-v2-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 13.266s
jic062-instruct-v17-g4-v2-mkmlizer: creating bucket guanaco-reward-models
jic062-instruct-v17-g4-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jic062-instruct-v17-g4-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jic062-instruct-v17-g4-v2_reward
jic062-instruct-v17-g4-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jic062-instruct-v17-g4-v2_reward/special_tokens_map.json
jic062-instruct-v17-g4-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jic062-instruct-v17-g4-v2_reward/config.json
jic062-instruct-v17-g4-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jic062-instruct-v17-g4-v2_reward/tokenizer_config.json
jic062-instruct-v17-g4-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jic062-instruct-v17-g4-v2_reward/merges.txt
jic062-instruct-v17-g4-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jic062-instruct-v17-g4-v2_reward/vocab.json
jic062-instruct-v17-g4-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jic062-instruct-v17-g4-v2_reward/tokenizer.json
jic062-instruct-v17-g4-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jic062-instruct-v17-g4-v2_reward/reward.tensors
Job jic062-instruct-v17-g4-v2-mkmlizer completed after 94.48s with status: succeeded
Stopping job with name jic062-instruct-v17-g4-v2-mkmlizer
Pipeline stage MKMLizer completed in 95.68s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.08s
Running pipeline stage ISVCDeployer
Creating inference service jic062-instruct-v17-g4-v2
Waiting for inference service jic062-instruct-v17-g4-v2 to be ready
Inference service jic062-instruct-v17-g4-v2 ready after 150.89581632614136s
Pipeline stage ISVCDeployer completed in 152.82s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3404972553253174s
Received healthy response to inference request in 1.4891481399536133s
Received healthy response to inference request in 1.4509401321411133s
Received healthy response to inference request in 1.390474557876587s
Received healthy response to inference request in 1.4994571208953857s
5 requests
0 failed requests
5th percentile: 1.4025676727294922
10th percentile: 1.4146607875823975
20th percentile: 1.438847017288208
30th percentile: 1.4585817337036133
40th percentile: 1.4738649368286132
50th percentile: 1.4891481399536133
60th percentile: 1.4932717323303222
70th percentile: 1.4973953247070313
80th percentile: 1.6676651477813722
90th percentile: 2.004081201553345
95th percentile: 2.1722892284393307
99th percentile: 2.30685564994812
mean time: 1.6341034412384032
Pipeline stage StressChecker completed in 9.23s
jic062-instruct-v17-g4_v2 status is now deployed due to DeploymentManager action
jic062-instruct-v17-g4_v2 status is now inactive due to auto deactivation removed underperforming models
jic062-instruct-v17-g4_v2 status is now torndown due to DeploymentManager action