submission_id: google-gemma-2-27b-it_v3
developer_uid: Jellywibble
alignment_samples: 0
best_of: 4
celo_rating: 1167.69
display_name: google-gemma-2-27b-it_v1
formatter: {'memory_template': "You are roleplaying as {bot_name} with a user. Be descriptive and creative in your responses while adhering to your persona. Here is {bot_name}'s Persona: {memory}\n", 'prompt_template': 'Example conversation: {prompt}\n<start_of_conversation>\n', 'bot_template': '{bot_name}: {message}<end_of_turn>\n', 'user_template': '<begin_of_turn>{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<end_of_turn>'], 'max_input_tokens': 512, 'best_of': 4, 'max_output_tokens': 64}
is_internal_developer: True
language_model: google/gemma-2-27b-it
max_input_tokens: 512
max_output_tokens: 64
model_architecture: Gemma2ForCausalLM
model_group: google/gemma-2-27b-it
model_name: google-gemma-2-27b-it_v1
model_num_parameters: 28731935232.0
model_repo: google/gemma-2-27b-it
model_size: 29B
num_battles: 115141
num_wins: 52604
propriety_score: 0.7378262173782622
propriety_total_count: 10001.0
ranking_group: single
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'}
reward_repo: ChaiML/gpt2_xl_pairwise_89m_step_347634
status: torndown
submission_type: basic
timestamp: 2024-07-14T00:38:33+00:00
us_pacific_date: 2024-07-13
win_ratio: 0.4568659295993608
Resubmit model
Running pipeline stage MKMLizer
Starting job with name google-gemma-2-27b-it-v3-mkmlizer
Waiting for job on google-gemma-2-27b-it-v3-mkmlizer to finish
google-gemma-2-27b-it-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
google-gemma-2-27b-it-v3-mkmlizer: ║ _____ __ __ ║
google-gemma-2-27b-it-v3-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
google-gemma-2-27b-it-v3-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
google-gemma-2-27b-it-v3-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
google-gemma-2-27b-it-v3-mkmlizer: ║ /___/ ║
google-gemma-2-27b-it-v3-mkmlizer: ║ ║
google-gemma-2-27b-it-v3-mkmlizer: ║ Version: 0.9.5.post2 ║
google-gemma-2-27b-it-v3-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
google-gemma-2-27b-it-v3-mkmlizer: ║ https://mk1.ai ║
google-gemma-2-27b-it-v3-mkmlizer: ║ ║
google-gemma-2-27b-it-v3-mkmlizer: ║ The license key for the current software has been verified as ║
google-gemma-2-27b-it-v3-mkmlizer: ║ belonging to: ║
google-gemma-2-27b-it-v3-mkmlizer: ║ ║
google-gemma-2-27b-it-v3-mkmlizer: ║ Chai Research Corp. ║
google-gemma-2-27b-it-v3-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
google-gemma-2-27b-it-v3-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
google-gemma-2-27b-it-v3-mkmlizer: ║ ║
google-gemma-2-27b-it-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
google-gemma-2-27b-it-v3-mkmlizer: Downloaded to shared memory in 81.641s
google-gemma-2-27b-it-v3-mkmlizer: quantizing model to /dev/shm/model_cache
google-gemma-2-27b-it-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
google-gemma-2-27b-it-v3-mkmlizer: quantized model in 84.858s
google-gemma-2-27b-it-v3-mkmlizer: Processed model google/gemma-2-27b-it in 166.500s
google-gemma-2-27b-it-v3-mkmlizer: creating bucket guanaco-mkml-models
google-gemma-2-27b-it-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
google-gemma-2-27b-it-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/google-gemma-2-27b-it-v3
google-gemma-2-27b-it-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/google-gemma-2-27b-it-v3/config.json
google-gemma-2-27b-it-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/google-gemma-2-27b-it-v3/special_tokens_map.json
google-gemma-2-27b-it-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/google-gemma-2-27b-it-v3/tokenizer_config.json
google-gemma-2-27b-it-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/google-gemma-2-27b-it-v3/tokenizer.model
google-gemma-2-27b-it-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/google-gemma-2-27b-it-v3/tokenizer.json
google-gemma-2-27b-it-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.2.safetensors s3://guanaco-mkml-models/google-gemma-2-27b-it-v3/flywheel_model.2.safetensors
google-gemma-2-27b-it-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/google-gemma-2-27b-it-v3/flywheel_model.0.safetensors
google-gemma-2-27b-it-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/google-gemma-2-27b-it-v3/flywheel_model.1.safetensors
google-gemma-2-27b-it-v3-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
google-gemma-2-27b-it-v3-mkmlizer: Loading 0: 0%| | 0/508 [00:00<?, ?it/s] Loading 0: 1%| | 4/508 [00:00<00:13, 36.00it/s] Loading 0: 3%|▎ | 14/508 [00:00<00:08, 60.69it/s] Loading 0: 4%|▍ | 20/508 [00:00<00:08, 57.17it/s] Loading 0: 6%|▌ | 28/508 [00:00<00:15, 30.01it/s] Loading 0: 6%|▋ | 33/508 [00:00<00:15, 30.24it/s] Loading 0: 7%|▋ | 37/508 [00:01<00:14, 31.79it/s] Loading 0: 9%|▉ | 47/508 [00:01<00:10, 43.92it/s] Loading 0: 10%|█ | 53/508 [00:01<00:09, 46.07it/s] Loading 0: 12%|█▏ | 59/508 [00:01<00:09, 46.74it/s] Loading 0: 13%|█▎ | 68/508 [00:01<00:08, 49.93it/s] Loading 0: 15%|█▍ | 74/508 [00:02<00:16, 26.95it/s] Loading 0: 16%|█▌ | 81/508 [00:02<00:13, 32.01it/s] Loading 0: 17%|█▋ | 87/508 [00:02<00:11, 36.24it/s] Loading 0: 19%|█▉ | 97/508 [00:02<00:08, 45.85it/s] Loading 0: 20%|██ | 103/508 [00:02<00:08, 47.69it/s] Loading 0: 21%|██▏ | 109/508 [00:02<00:08, 48.26it/s] Loading 0: 23%|██▎ | 118/508 [00:03<00:12, 31.70it/s] Loading 0: 24%|██▍ | 124/508 [00:03<00:11, 34.05it/s] Loading 0: 25%|██▌ | 129/508 [00:03<00:10, 35.75it/s] Loading 0: 27%|██▋ | 136/508 [00:03<00:09, 40.59it/s] Loading 0: 29%|██▊ | 146/508 [00:03<00:07, 50.10it/s] Loading 0: 30%|██▉ | 152/508 [00:03<00:06, 51.04it/s] Loading 0: 31%|███ | 158/508 [00:03<00:06, 50.88it/s] Loading 0: 34%|███▎ | 171/508 [00:04<00:08, 37.57it/s] Loading 0: 35%|███▍ | 176/508 [00:04<00:09, 36.07it/s] Loading 0: 36%|███▌ | 181/508 [00:04<00:09, 34.90it/s] Loading 0: 38%|███▊ | 191/508 [00:04<00:07, 44.58it/s] Loading 0: 38%|███▊ | 191/508 [00:25<00:07, 44.58it/s] Loading 0: 38%|███▊ | 192/508 [00:25<05:47, 1.10s/it] Loading 0: 40%|███▉ | 202/508 [00:25<03:14, 1.58it/s] Loading 0: 42%|████▏ | 211/508 [00:25<02:03, 2.41it/s] Loading 0: 43%|████▎ | 218/508 [00:26<01:32, 3.13it/s] Loading 0: 44%|████▍ | 224/508 [00:26<01:08, 4.14it/s] Loading 0: 46%|████▌ | 234/508 [00:26<00:42, 6.46it/s] Loading 0: 47%|████▋ | 240/508 [00:26<00:32, 8.25it/s] Loading 0: 48%|████▊ | 246/508 [00:26<00:24, 10.53it/s] Loading 0: 50%|█████ | 255/508 [00:26<00:16, 15.16it/s] Loading 0: 52%|█████▏ | 262/508 [00:27<00:16, 14.94it/s] Loading 0: 53%|█████▎ | 268/508 [00:27<00:13, 18.34it/s] Loading 0: 55%|█████▍ | 278/508 [00:27<00:08, 25.92it/s] Loading 0: 56%|█████▌ | 284/508 [00:27<00:07, 29.65it/s] Loading 0: 57%|█████▋ | 290/508 [00:27<00:06, 33.13it/s] Loading 0: 59%|█████▉ | 300/508 [00:27<00:04, 42.60it/s] Loading 0: 60%|██████ | 307/508 [00:27<00:04, 46.06it/s] Loading 0: 62%|██████▏ | 314/508 [00:28<00:06, 31.47it/s] Loading 0: 63%|██████▎ | 319/508 [00:28<00:06, 31.40it/s] Loading 0: 64%|██████▍ | 324/508 [00:28<00:05, 31.49it/s] Loading 0: 66%|██████▌ | 334/508 [00:28<00:04, 41.72it/s] Loading 0: 68%|██████▊ | 344/508 [00:28<00:03, 50.63it/s] Loading 0: 69%|██████▉ | 350/508 [00:29<00:03, 51.04it/s] Loading 0: 71%|███████ | 359/508 [00:29<00:04, 33.32it/s] Loading 0: 72%|███████▏ | 364/508 [00:29<00:04, 35.50it/s] Loading 0: 73%|███████▎ | 369/508 [00:29<00:04, 34.44it/s] Loading 0: 74%|███████▍ | 378/508 [00:29<00:03, 42.59it/s] Loading 0: 76%|███████▋ | 388/508 [00:30<00:02, 51.42it/s] Loading 0: 78%|███████▊ | 394/508 [00:30<00:02, 51.93it/s] Loading 0: 79%|███████▊ | 400/508 [00:30<00:02, 50.37it/s] Loading 0: 80%|███████▉ | 406/508 [00:30<00:03, 29.30it/s] Loading 0: 81%|████████ | 411/508 [00:30<00:02, 32.35it/s] Loading 0: 83%|████████▎ | 421/508 [00:30<00:02, 42.57it/s] Loading 0: 84%|████████▍ | 427/508 [00:31<00:01, 45.07it/s] Loading 0: 85%|████████▌ | 433/508 [00:31<00:01, 46.32it/s] Loading 0: 87%|████████▋ | 443/508 [00:31<00:01, 55.39it/s] Loading 0: 89%|████████▊ | 450/508 [00:31<00:01, 56.48it/s] Loading 0: 89%|████████▉ | 451/508 [00:52<00:01, 56.48it/s] Loading 0: 89%|████████▉ | 452/508 [00:52<01:01, 1.10s/it] Loading 0: 90%|████████▉ | 457/508 [00:52<00:41, 1.21it/s] Loading 0: 91%|█████████ | 462/508 [00:52<00:27, 1.66it/s] Loading 0: 92%|█████████▏| 467/508 [00:53<00:18, 2.27it/s] Loading 0: 94%|█████████▍| 477/508 [00:53<00:07, 4.01it/s] Loading 0: 96%|█████████▌| 487/508 [00:53<00:03, 6.36it/s] Loading 0: 97%|█████████▋| 493/508 [00:53<00:01, 8.15it/s] Loading 0: 99%|█████████▉| 502/508 [00:53<00:00, 10.07it/s] Loading 0: 100%|█████████▉| 507/508 [00:54<00:00, 12.12it/s] /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:950: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
google-gemma-2-27b-it-v3-mkmlizer: warnings.warn(
google-gemma-2-27b-it-v3-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:778: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
google-gemma-2-27b-it-v3-mkmlizer: warnings.warn(
google-gemma-2-27b-it-v3-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.
google-gemma-2-27b-it-v3-mkmlizer: warnings.warn(
google-gemma-2-27b-it-v3-mkmlizer: Downloading shards: 0%| | 0/2 [00:00<?, ?it/s] Downloading shards: 50%|█████ | 1/2 [00:07<00:07, 7.74s/it] Downloading shards: 100%|██████████| 2/2 [00:10<00:00, 4.65s/it] Downloading shards: 100%|██████████| 2/2 [00:10<00:00, 5.11s/it]
google-gemma-2-27b-it-v3-mkmlizer: Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.57it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 2.59it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 2.36it/s]
google-gemma-2-27b-it-v3-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
google-gemma-2-27b-it-v3-mkmlizer: Saving duration: 2.356s
google-gemma-2-27b-it-v3-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 15.310s
google-gemma-2-27b-it-v3-mkmlizer: creating bucket guanaco-reward-models
google-gemma-2-27b-it-v3-mkmlizer: Bucket 's3://guanaco-reward-models/' created
google-gemma-2-27b-it-v3-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/google-gemma-2-27b-it-v3_reward
google-gemma-2-27b-it-v3-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/google-gemma-2-27b-it-v3_reward/config.json
google-gemma-2-27b-it-v3-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/google-gemma-2-27b-it-v3_reward/tokenizer_config.json
google-gemma-2-27b-it-v3-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/google-gemma-2-27b-it-v3_reward/merges.txt
google-gemma-2-27b-it-v3-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/google-gemma-2-27b-it-v3_reward/special_tokens_map.json
google-gemma-2-27b-it-v3-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/google-gemma-2-27b-it-v3_reward/tokenizer.json
google-gemma-2-27b-it-v3-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/google-gemma-2-27b-it-v3_reward/vocab.json
Job google-gemma-2-27b-it-v3-mkmlizer completed after 261.56s with status: succeeded
Stopping job with name google-gemma-2-27b-it-v3-mkmlizer
Pipeline stage MKMLizer completed in 262.43s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service google-gemma-2-27b-it-v3
Waiting for inference service google-gemma-2-27b-it-v3 to be ready
Failed to get response for submission blend_meral_2024-07-13: ('http://mistralai-mixtral-8x7b-3473-v86-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:50640->127.0.0.1:8080: read: connection reset by peer\n')
Inference service google-gemma-2-27b-it-v3 ready after 100.45689940452576s
Pipeline stage ISVCDeployer completed in 107.55s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.5726592540740967s
Received healthy response to inference request in 2.583028554916382s
Received healthy response to inference request in 2.483349323272705s
Received healthy response to inference request in 3.2021548748016357s
Received healthy response to inference request in 3.1752748489379883s
5 requests
0 failed requests
5th percentile: 2.5032851696014404
10th percentile: 2.523221015930176
20th percentile: 2.5630927085876465
30th percentile: 2.701477813720703
40th percentile: 2.9383763313293456
50th percentile: 3.1752748489379883
60th percentile: 3.186026859283447
70th percentile: 3.1967788696289063
80th percentile: 3.276255750656128
90th percentile: 3.4244575023651125
95th percentile: 3.4985583782196046
99th percentile: 3.557839078903198
mean time: 3.0032933712005616
Pipeline stage StressChecker completed in 15.63s
google-gemma-2-27b-it_v3 status is now deployed due to DeploymentManager action
google-gemma-2-27b-it_v3 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of google-gemma-2-27b-it_v3
Running pipeline stage ISVCDeleter
Checking if service google-gemma-2-27b-it-v3 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 4.25s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key google-gemma-2-27b-it-v3/config.json from bucket guanaco-mkml-models
Deleting key google-gemma-2-27b-it-v3/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key google-gemma-2-27b-it-v3/flywheel_model.1.safetensors from bucket guanaco-mkml-models
Deleting key google-gemma-2-27b-it-v3/flywheel_model.2.safetensors from bucket guanaco-mkml-models
Deleting key google-gemma-2-27b-it-v3/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key google-gemma-2-27b-it-v3/tokenizer.json from bucket guanaco-mkml-models
Deleting key google-gemma-2-27b-it-v3/tokenizer.model from bucket guanaco-mkml-models
Deleting key google-gemma-2-27b-it-v3/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key google-gemma-2-27b-it-v3_reward/config.json from bucket guanaco-reward-models
Deleting key google-gemma-2-27b-it-v3_reward/merges.txt from bucket guanaco-reward-models
Deleting key google-gemma-2-27b-it-v3_reward/reward.tensors from bucket guanaco-reward-models
Deleting key google-gemma-2-27b-it-v3_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key google-gemma-2-27b-it-v3_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key google-gemma-2-27b-it-v3_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key google-gemma-2-27b-it-v3_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 8.42s
google-gemma-2-27b-it_v3 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics