submission_id: intervitens-mini-magnum-_5180_v1
developer_uid: alpin
best_of: 4
celo_rating: 1200.46
display_name: magnum-12b-v1
family_friendly_score: 0.0
formatter: {'memory_template': "<s>{bot_name}'s Persona: {memory}\\\n\\\n", 'prompt_template': '{prompt} [/INST]', 'bot_template': '{bot_name}: {message}</s>', 'user_template': '[INST] {user_name}: {message} [/INST]', 'response_template': '{bot_name}:', 'truncate_by_message': False}
generation_params: {'temperature': 1.15, 'top_p': 1.0, 'min_p': 0.07, 'top_k': 1024, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 1024, 'best_of': 4, 'max_output_tokens': 64, 'reward_max_token_input': 1024}
is_internal_developer: False
language_model: intervitens/mini-magnum-12b-v1.1
max_input_tokens: 1024
max_output_tokens: 64
model_architecture: MistralForCausalLM
model_group: intervitens/mini-magnum-
model_name: magnum-12b-v1
model_num_parameters: 12772080640.0
model_repo: intervitens/mini-magnum-12b-v1.1
model_size: 13B
num_battles: 11473
num_wins: 5658
ranking_group: single
reward_formatter: {'bot_template': '<|im_start|>assistant\n{bot_name}: {message}<|im_end|>\n', 'memory_template': '<|im_start|>system\n{memory}<|im_end|>\n', 'prompt_template': '<|im_start|>user\n{prompt}<|im_end|>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '<|im_start|>user\n{user_name}: {message}<|im_end|>\n'}
reward_repo: Jellywibble/gpt2_xl_pairwise_89m_step_347634
status: torndown
submission_type: basic
timestamp: 2024-07-27T00:04:38+00:00
us_pacific_date: 2024-07-26
win_ratio: 0.4931578488625468
Download Preference Data
Resubmit model
Running pipeline stage MKMLizer
Starting job with name intervitens-mini-magnum-5180-v1-mkmlizer
Waiting for job on intervitens-mini-magnum-5180-v1-mkmlizer to finish
intervitens-mini-magnum-5180-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
intervitens-mini-magnum-5180-v1-mkmlizer: ║ _____ __ __ ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ /___/ ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ Version: 0.9.7 ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ https://mk1.ai ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ The license key for the current software has been verified as ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ belonging to: ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ Chai Research Corp. ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
intervitens-mini-magnum-5180-v1-mkmlizer: ║ ║
intervitens-mini-magnum-5180-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
intervitens-mini-magnum-5180-v1-mkmlizer: Downloaded to shared memory in 35.401s
intervitens-mini-magnum-5180-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpvt32cp39, device:0
intervitens-mini-magnum-5180-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
intervitens-mini-magnum-5180-v1-mkmlizer: quantized model in 37.010s
intervitens-mini-magnum-5180-v1-mkmlizer: Processed model intervitens/mini-magnum-12b-v1.1 in 72.410s
intervitens-mini-magnum-5180-v1-mkmlizer: creating bucket guanaco-mkml-models
intervitens-mini-magnum-5180-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
intervitens-mini-magnum-5180-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/intervitens-mini-magnum-5180-v1
intervitens-mini-magnum-5180-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/intervitens-mini-magnum-5180-v1/config.json
intervitens-mini-magnum-5180-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/intervitens-mini-magnum-5180-v1/special_tokens_map.json
intervitens-mini-magnum-5180-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/intervitens-mini-magnum-5180-v1/tokenizer_config.json
intervitens-mini-magnum-5180-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/intervitens-mini-magnum-5180-v1/tokenizer.json
intervitens-mini-magnum-5180-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/intervitens-mini-magnum-5180-v1/flywheel_model.0.safetensors
intervitens-mini-magnum-5180-v1-mkmlizer: loading reward model from Jellywibble/gpt2_xl_pairwise_89m_step_347634
intervitens-mini-magnum-5180-v1-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 1%|▏ | 5/363 [00:00<00:11, 31.50it/s] Loading 0: 3%|▎ | 12/363 [00:00<00:07, 49.66it/s] Loading 0: 5%|▍ | 18/363 [00:00<00:06, 49.49it/s] Loading 0: 7%|▋ | 24/363 [00:00<00:08, 41.20it/s] Loading 0: 9%|▊ | 31/363 [00:00<00:07, 46.89it/s] Loading 0: 10%|█ | 37/363 [00:00<00:07, 44.17it/s] Loading 0: 12%|█▏ | 42/363 [00:00<00:07, 42.15it/s] Loading 0: 13%|█▎ | 49/363 [00:01<00:06, 46.61it/s] Loading 0: 15%|█▌ | 55/363 [00:01<00:07, 43.46it/s] Loading 0: 17%|█▋ | 60/363 [00:01<00:06, 44.93it/s] Loading 0: 18%|█▊ | 65/363 [00:01<00:10, 28.78it/s] Loading 0: 20%|█▉ | 72/363 [00:01<00:08, 34.88it/s] Loading 0: 21%|██ | 77/363 [00:01<00:07, 37.19it/s] Loading 0: 23%|██▎ | 82/363 [00:02<00:08, 33.40it/s] Loading 0: 25%|██▍ | 90/363 [00:02<00:06, 41.56it/s] Loading 0: 26%|██▋ | 96/363 [00:02<00:06, 40.58it/s] Loading 0: 28%|██▊ | 101/363 [00:02<00:06, 39.47it/s] Loading 0: 30%|██▉ | 108/363 [00:02<00:05, 45.88it/s] Loading 0: 31%|███ | 113/363 [00:02<00:06, 37.81it/s] Loading 0: 33%|███▎ | 118/363 [00:02<00:06, 37.03it/s] Loading 0: 34%|███▍ | 125/363 [00:03<00:05, 43.61it/s] Loading 0: 36%|███▌ | 130/363 [00:03<00:05, 42.40it/s] Loading 0: 37%|███▋ | 135/363 [00:03<00:05, 42.62it/s] Loading 0: 39%|███▉ | 141/363 [00:03<00:05, 41.74it/s] Loading 0: 40%|████ | 146/363 [00:03<00:07, 29.53it/s] Loading 0: 41%|████▏ | 150/363 [00:03<00:07, 29.46it/s] Loading 0: 43%|████▎ | 156/363 [00:04<00:05, 35.44it/s] Loading 0: 44%|████▍ | 161/363 [00:04<00:05, 37.39it/s] Loading 0: 46%|████▌ | 166/363 [00:04<00:05, 39.03it/s] Loading 0: 47%|████▋ | 172/363 [00:04<00:04, 39.27it/s] Loading 0: 49%|████▉ | 177/363 [00:04<00:04, 38.06it/s] Loading 0: 50%|█████ | 183/363 [00:04<00:04, 43.03it/s] Loading 0: 52%|█████▏ | 188/363 [00:04<00:04, 41.71it/s] Loading 0: 53%|█████▎ | 193/363 [00:04<00:03, 43.24it/s] Loading 0: 55%|█████▍ | 199/363 [00:05<00:03, 42.08it/s] Loading 0: 56%|█████▌ | 204/363 [00:05<00:03, 41.68it/s] Loading 0: 58%|█████▊ | 210/363 [00:05<00:03, 45.78it/s] Loading 0: 59%|█████▉ | 215/363 [00:05<00:03, 41.94it/s] Loading 0: 61%|██████ | 220/363 [00:05<00:03, 42.62it/s] Loading 0: 62%|██████▏ | 225/363 [00:05<00:05, 26.42it/s] Loading 0: 63%|██████▎ | 230/363 [00:06<00:04, 28.80it/s] Loading 0: 65%|██████▌ | 237/363 [00:06<00:03, 36.50it/s] Loading 0: 67%|██████▋ | 242/363 [00:06<00:03, 37.88it/s] Loading 0: 68%|██████▊ | 247/363 [00:06<00:02, 39.10it/s] Loading 0: 69%|██████▉ | 252/363 [00:06<00:02, 40.74it/s] Loading 0: 71%|███████ | 257/363 [00:06<00:03, 35.10it/s] Loading 0: 73%|███████▎ | 264/363 [00:06<00:02, 42.39it/s] Loading 0: 74%|███████▍ | 269/363 [00:06<00:02, 42.69it/s] Loading 0: 76%|███████▌ | 275/363 [00:07<00:02, 39.49it/s] Loading 0: 78%|███████▊ | 283/363 [00:07<00:01, 47.07it/s] Loading 0: 80%|███████▉ | 289/363 [00:07<00:01, 43.90it/s] Loading 0: 81%|████████ | 294/363 [00:07<00:01, 41.77it/s] Loading 0: 83%|████████▎ | 300/363 [00:07<00:01, 46.02it/s] Loading 0: 84%|████████▍ | 305/363 [00:14<00:22, 2.60it/s] Loading 0: 85%|████████▌ | 309/363 [00:14<00:16, 3.32it/s] Loading 0: 86%|████████▌ | 313/363 [00:14<00:11, 4.21it/s] Loading 0: 88%|████████▊ | 319/363 [00:14<00:07, 6.22it/s] Loading 0: 89%|████████▉ | 323/363 [00:14<00:05, 7.86it/s] Loading 0: 91%|█████████ | 329/363 [00:15<00:03, 11.18it/s] Loading 0: 92%|█████████▏| 335/363 [00:15<00:01, 14.73it/s] Loading 0: 94%|█████████▎| 340/363 [00:15<00:01, 18.03it/s] Loading 0: 96%|█████████▌| 347/363 [00:15<00:00, 24.17it/s] Loading 0: 97%|█████████▋| 352/363 [00:15<00:00, 27.93it/s] Loading 0: 98%|█████████▊| 357/363 [00:15<00:00, 28.23it/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.
intervitens-mini-magnum-5180-v1-mkmlizer: warnings.warn(
intervitens-mini-magnum-5180-v1-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.
intervitens-mini-magnum-5180-v1-mkmlizer: warnings.warn(
intervitens-mini-magnum-5180-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
intervitens-mini-magnum-5180-v1-mkmlizer: Saving duration: 1.429s
intervitens-mini-magnum-5180-v1-mkmlizer: Processed model Jellywibble/gpt2_xl_pairwise_89m_step_347634 in 11.023s
intervitens-mini-magnum-5180-v1-mkmlizer: creating bucket guanaco-reward-models
intervitens-mini-magnum-5180-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
intervitens-mini-magnum-5180-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/intervitens-mini-magnum-5180-v1_reward
intervitens-mini-magnum-5180-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/intervitens-mini-magnum-5180-v1_reward/config.json
intervitens-mini-magnum-5180-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/intervitens-mini-magnum-5180-v1_reward/tokenizer_config.json
intervitens-mini-magnum-5180-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/intervitens-mini-magnum-5180-v1_reward/special_tokens_map.json
intervitens-mini-magnum-5180-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/intervitens-mini-magnum-5180-v1_reward/merges.txt
intervitens-mini-magnum-5180-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/intervitens-mini-magnum-5180-v1_reward/vocab.json
intervitens-mini-magnum-5180-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/intervitens-mini-magnum-5180-v1_reward/tokenizer.json
intervitens-mini-magnum-5180-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/intervitens-mini-magnum-5180-v1_reward/reward.tensors
Job intervitens-mini-magnum-5180-v1-mkmlizer completed after 115.14s with status: succeeded
Stopping job with name intervitens-mini-magnum-5180-v1-mkmlizer
Pipeline stage MKMLizer completed in 116.20s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service intervitens-mini-magnum-5180-v1
Waiting for inference service intervitens-mini-magnum-5180-v1 to be ready
Inference service intervitens-mini-magnum-5180-v1 ready after 81.06340956687927s
Pipeline stage ISVCDeployer completed in 82.63s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.531583786010742s
Received healthy response to inference request in 1.6261050701141357s
Received healthy response to inference request in 1.777008295059204s
Received healthy response to inference request in 1.801353931427002s
Received healthy response to inference request in 1.747403860092163s
5 requests
0 failed requests
5th percentile: 1.6503648281097412
10th percentile: 1.6746245861053466
20th percentile: 1.7231441020965577
30th percentile: 1.7533247470855713
40th percentile: 1.7651665210723877
50th percentile: 1.777008295059204
60th percentile: 1.7867465496063233
70th percentile: 1.7964848041534425
80th percentile: 1.94739990234375
90th percentile: 2.239491844177246
95th percentile: 2.385537815093994
99th percentile: 2.5023745918273925
mean time: 1.8966909885406493
Pipeline stage StressChecker completed in 10.18s
intervitens-mini-magnum-_5180_v1 status is now deployed due to DeploymentManager action
intervitens-mini-magnum-_5180_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of intervitens-mini-magnum-_5180_v1
Running pipeline stage ISVCDeleter
Checking if service intervitens-mini-magnum-5180-v1 is running
Tearing down inference service intervitens-mini-magnum-5180-v1
Service intervitens-mini-magnum-5180-v1 has been torndown
Pipeline stage ISVCDeleter completed in 5.07s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key intervitens-mini-magnum-5180-v1/config.json from bucket guanaco-mkml-models
Deleting key intervitens-mini-magnum-5180-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key intervitens-mini-magnum-5180-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key intervitens-mini-magnum-5180-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key intervitens-mini-magnum-5180-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key intervitens-mini-magnum-5180-v1_reward/config.json from bucket guanaco-reward-models
Deleting key intervitens-mini-magnum-5180-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key intervitens-mini-magnum-5180-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key intervitens-mini-magnum-5180-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key intervitens-mini-magnum-5180-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key intervitens-mini-magnum-5180-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key intervitens-mini-magnum-5180-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 6.46s
intervitens-mini-magnum-_5180_v1 status is now torndown due to DeploymentManager action