jic062-instruct

developer_uid: chace9580

submission_id: jic062-instruct_v8

model_name: jic062-instruct_v8

model_group: jic062/instruct

status: torndown

timestamp: 2024-07-29T06:17:32+00:00

num_battles: 13097

num_wins: 6361

celo_rating: 1192.32

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: 16

max_input_tokens: 512

max_output_tokens: 64

display_name: jic062-instruct_v8

is_internal_developer: False

language_model: jic062/instruct

model_size: 8B

ranking_group: single

us_pacific_date: 2024-07-28

win_ratio: 0.48568374436893946

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|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64, 'reward_max_token_input': 256}

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}

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-v8-mkmlizer
Waiting for job on jic062-instruct-v8-mkmlizer to finish
jic062-instruct-v8-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-instruct-v8-mkmlizer: ║     _____            __           __                                ║
jic062-instruct-v8-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-instruct-v8-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-instruct-v8-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-instruct-v8-mkmlizer: ║       /___/                                                         ║
jic062-instruct-v8-mkmlizer: ║                                                                     ║
jic062-instruct-v8-mkmlizer: ║  Version: 0.9.7                                                     ║
jic062-instruct-v8-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-instruct-v8-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-instruct-v8-mkmlizer: ║                                                                     ║
jic062-instruct-v8-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-instruct-v8-mkmlizer: ║  belonging to:                                                      ║
jic062-instruct-v8-mkmlizer: ║                                                                     ║
jic062-instruct-v8-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-instruct-v8-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-instruct-v8-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-instruct-v8-mkmlizer: ║                                                                     ║
jic062-instruct-v8-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-instruct-v8-mkmlizer: Downloaded to shared memory in 29.509s
jic062-instruct-v8-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmprysvvl3x, device:0
jic062-instruct-v8-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-instruct-v8-mkmlizer: creating bucket guanaco-mkml-models
jic062-instruct-v8-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-instruct-v8-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-instruct-v8
jic062-instruct-v8-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-instruct-v8/config.json
jic062-instruct-v8-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-instruct-v8/special_tokens_map.json
jic062-instruct-v8-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-instruct-v8/tokenizer_config.json
jic062-instruct-v8-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-instruct-v8/tokenizer.json
jic062-instruct-v8-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-instruct-v8/flywheel_model.0.safetensors
jic062-instruct-v8-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
jic062-instruct-v8-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 52.57it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:04, 68.68it/s]
Loading 0:   9%|▉         | 26/291 [00:00<00:03, 81.05it/s]
Loading 0:  12%|█▏        | 35/291 [00:00<00:03, 81.23it/s]
Loading 0:  17%|█▋        | 49/291 [00:00<00:02, 86.57it/s]
Loading 0:  21%|██        | 60/291 [00:00<00:02, 92.86it/s]
Loading 0:  24%|██▍       | 70/291 [00:00<00:02, 83.62it/s]
Loading 0:  27%|██▋       | 79/291 [00:00<00:02, 83.20it/s]
Loading 0:  30%|███       | 88/291 [00:02<00:09, 22.38it/s]
Loading 0:  33%|███▎      | 97/291 [00:02<00:06, 28.62it/s]
Loading 0:  36%|███▋      | 106/291 [00:02<00:05, 35.54it/s]
Loading 0:  40%|███▉      | 115/291 [00:02<00:04, 42.47it/s]
Loading 0:  43%|████▎     | 124/291 [00:02<00:03, 50.18it/s]
Loading 0:  46%|████▌     | 133/291 [00:02<00:02, 57.61it/s]
Loading 0:  49%|████▉     | 142/291 [00:02<00:02, 63.52it/s]
Loading 0:  52%|█████▏    | 152/291 [00:02<00:01, 71.83it/s]
Loading 0:  57%|█████▋    | 166/291 [00:02<00:01, 79.49it/s]
Loading 0:  60%|██████    | 175/291 [00:03<00:01, 82.02it/s]
Loading 0:  63%|██████▎   | 184/291 [00:03<00:01, 81.00it/s]
Loading 0:  66%|██████▋   | 193/291 [00:04<00:04, 23.76it/s]
Loading 0:  69%|██████▉   | 202/291 [00:04<00:02, 30.08it/s]
Loading 0:  73%|███████▎  | 211/291 [00:04<00:02, 36.80it/s]
Loading 0:  76%|███████▌  | 220/291 [00:04<00:01, 44.35it/s]
Loading 0:  79%|███████▉  | 230/291 [00:04<00:01, 53.84it/s]
Loading 0:  82%|████████▏ | 239/291 [00:04<00:00, 60.79it/s]
Loading 0:  85%|████████▌ | 248/291 [00:04<00:00, 66.60it/s]
Loading 0:  88%|████████▊ | 257/291 [00:04<00:00, 69.12it/s]
Loading 0:  92%|█████████▏| 267/291 [00:05<00:00, 76.42it/s]
Loading 0:  95%|█████████▍| 276/291 [00:05<00:00, 79.23it/s]
Loading 0:  98%|█████████▊| 285/291 [00:05<00:00, 80.84it/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-v8-mkmlizer:   warnings.warn(
jic062-instruct-v8-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-v8-mkmlizer:   warnings.warn(
jic062-instruct-v8-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-v8-mkmlizer:   warnings.warn(
jic062-instruct-v8-mkmlizer: 
Downloading shards:   0%|          | 0/2 [00:00<?, ?it/s]
Downloading shards:  50%|█████     | 1/2 [00:05<00:05,  5.51s/it]
Downloading shards: 100%|██████████| 2/2 [00:08<00:00,  3.97s/it]
Downloading shards: 100%|██████████| 2/2 [00:08<00:00,  4.20s/it]
jic062-instruct-v8-mkmlizer: 
Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]
Loading checkpoint shards:  50%|█████     | 1/2 [00:00<00:00,  2.36it/s]
Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  3.87it/s]
Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  3.53it/s]
jic062-instruct-v8-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jic062-instruct-v8-mkmlizer: Saving duration: 1.365s
jic062-instruct-v8-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 13.365s
jic062-instruct-v8-mkmlizer: creating bucket guanaco-reward-models
jic062-instruct-v8-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jic062-instruct-v8-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jic062-instruct-v8_reward
jic062-instruct-v8-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jic062-instruct-v8_reward/tokenizer_config.json
jic062-instruct-v8-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jic062-instruct-v8_reward/config.json
jic062-instruct-v8-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jic062-instruct-v8_reward/special_tokens_map.json
jic062-instruct-v8-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jic062-instruct-v8_reward/merges.txt
jic062-instruct-v8-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jic062-instruct-v8_reward/vocab.json
jic062-instruct-v8-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jic062-instruct-v8_reward/tokenizer.json
jic062-instruct-v8-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jic062-instruct-v8_reward/reward.tensors
Job jic062-instruct-v8-mkmlizer completed after 106.02s with status: succeeded
Stopping job with name jic062-instruct-v8-mkmlizer
Pipeline stage MKMLizer completed in 107.21s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service jic062-instruct-v8
Waiting for inference service jic062-instruct-v8 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 jic062-instruct-v8 ready after 100.80276489257812s
Pipeline stage ISVCDeployer completed in 102.48s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2484686374664307s
Received healthy response to inference request in 1.3938636779785156s
Received healthy response to inference request in 1.3888239860534668s
Received healthy response to inference request in 1.3894388675689697s
Received healthy response to inference request in 1.3931894302368164s
5 requests
0 failed requests
5th percentile: 1.3889469623565673
10th percentile: 1.3890699386596679
20th percentile: 1.3893158912658692
30th percentile: 1.390188980102539
40th percentile: 1.3916892051696776
50th percentile: 1.3931894302368164
60th percentile: 1.3934591293334961
70th percentile: 1.3937288284301759
80th percentile: 1.5647846698760988
90th percentile: 1.9066266536712648
95th percentile: 2.0775476455688473
99th percentile: 2.214284439086914
mean time: 1.5627569198608398
Pipeline stage StressChecker completed in 8.53s
jic062-instruct_v8 status is now deployed due to DeploymentManager action
jic062-instruct_v8 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of jic062-instruct_v8
Running pipeline stage ISVCDeleter
Checking if service jic062-instruct-v8 is running
Tearing down inference service jic062-instruct-v8
Service jic062-instruct-v8 has been torndown
Pipeline stage ISVCDeleter completed in 4.83s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key jic062-instruct-v8/config.json from bucket guanaco-mkml-models
Deleting key jic062-instruct-v8/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key jic062-instruct-v8/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key jic062-instruct-v8/tokenizer.json from bucket guanaco-mkml-models
Deleting key jic062-instruct-v8/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key jic062-instruct-v8_reward/config.json from bucket guanaco-reward-models
Deleting key jic062-instruct-v8_reward/merges.txt from bucket guanaco-reward-models
Deleting key jic062-instruct-v8_reward/reward.tensors from bucket guanaco-reward-models
Deleting key jic062-instruct-v8_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key jic062-instruct-v8_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key jic062-instruct-v8_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key jic062-instruct-v8_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.47s
jic062-instruct_v8 status is now torndown due to DeploymentManager action