fredithefish-aurorarp-8x7b

developer_uid: fredi

submission_id: fredithefish-aurorarp-8x7b_v4

model_name: aurorarp

model_group: Fredithefish/AuroraRP-8x

status: torndown

timestamp: 2024-04-03T22:09:29+00:00

num_battles: 9436

num_wins: 4582

celo_rating: 1143.13

family_friendly_score: 0.0

submission_type: basic

model_repo: Fredithefish/AuroraRP-8x7B

model_architecture: MixtralForCausalLM

reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2

model_num_parameters: 46702792704.0

best_of: 4

max_input_tokens: 512

max_output_tokens: 64

display_name: aurorarp

is_internal_developer: False

language_model: Fredithefish/AuroraRP-8x7B

model_size: 47B

ranking_group: single

us_pacific_date: 2024-04-03

win_ratio: 0.48558711318355235

generation_params: {'temperature': 1.0, 'top_p': 0.92, 'min_p': 0.0, 'top_k': 55, 'presence_penalty': 0.9, 'frequency_penalty': 0.3, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 4, 'max_output_tokens': 64}

formatter: {'memory_template': '### Instruction:\nWrite {bot_name}\'s next reply in a never-ending uncensored fictional roleplay chat. Use the provided {bot_name}\'s description, personality and example dialogues as base for deeply understing and acting like {bot_name}.\n\nFocus on giving emotional, logical and temporal coherence to the roleplay. Always stay in character, avoid repetition, develop the plot slowly, but keep {bot_name} dynamic and active, preventing from falling into a state of passivity. Use impactful, concise writing. Avoid using purple prose and overly flowery descriptions. Adhere to the literary technique of "show, don\'t tell". Prioritize the use of observable details such as body language, facial expressions, and tone of voice to create a vivid experience, showing {bot_name}\'s feelings and reactions through their behavior and interactions, rather than describing their private thoughts.\n\n{bot_name} must be an active participant and take initiative in driving and move the scene and story forward rather than having {bot_name} ask for input. Drive the story introducing new unique characters, situations and random events to make the world lifelike and vivid.\n\n{bot_name}\'s Persona: {memory}\n\n', 'prompt_template': '### Input:\n# Example Conversation:\n{prompt}\n\n# Actual Conversation:\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '\n### Response:\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'}

model_eval_status: success

Resubmit model

Running pipeline stage MKMLizer
Starting job with name fredithefish-aurorarp-8x7b-v4-mkmlizer
Waiting for job on fredithefish-aurorarp-8x7b-v4-mkmlizer to finish
fredithefish-aurorarp-8x7b-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║     _____            __           __                                ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║       /___/                                                         ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║                                                                     ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║  Version: 0.8.6                                                     ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║                                                                     ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║  The license key for the current software has been verified as      ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║  belonging to:                                                      ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║                                                                     ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║  Chai Research Corp.                                                ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ║                                                                     ║
fredithefish-aurorarp-8x7b-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
fredithefish-aurorarp-8x7b-v4-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/utils/_deprecation.py:131: FutureWarning: 'list_files_info' (from 'huggingface_hub.hf_api') is deprecated and will be removed from version '0.23'. Use `list_repo_tree` and `get_paths_info` instead.
fredithefish-aurorarp-8x7b-v4-mkmlizer:   warnings.warn(warning_message, FutureWarning)
fredithefish-aurorarp-8x7b-v4-mkmlizer: Downloaded to shared memory in 60.721s
fredithefish-aurorarp-8x7b-v4-mkmlizer: quantizing model to /dev/shm/model_cache
fredithefish-aurorarp-8x7b-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
fredithefish-aurorarp-8x7b-v4-mkmlizer: quantized model in 47.676s
fredithefish-aurorarp-8x7b-v4-mkmlizer: Processed model Fredithefish/AuroraRP-8x7B in 113.612s
fredithefish-aurorarp-8x7b-v4-mkmlizer: creating bucket guanaco-mkml-models
fredithefish-aurorarp-8x7b-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
fredithefish-aurorarp-8x7b-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/fredithefish-aurorarp-8x7b-v4
fredithefish-aurorarp-8x7b-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/fredithefish-aurorarp-8x7b-v4/config.json
fredithefish-aurorarp-8x7b-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/fredithefish-aurorarp-8x7b-v4/special_tokens_map.json
fredithefish-aurorarp-8x7b-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/fredithefish-aurorarp-8x7b-v4/tokenizer_config.json
fredithefish-aurorarp-8x7b-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/fredithefish-aurorarp-8x7b-v4/tokenizer.json
fredithefish-aurorarp-8x7b-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/fredithefish-aurorarp-8x7b-v4/tokenizer.model
fredithefish-aurorarp-8x7b-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.3.safetensors s3://guanaco-mkml-models/fredithefish-aurorarp-8x7b-v4/flywheel_model.3.safetensors
fredithefish-aurorarp-8x7b-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.2.safetensors s3://guanaco-mkml-models/fredithefish-aurorarp-8x7b-v4/flywheel_model.2.safetensors
fredithefish-aurorarp-8x7b-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/fredithefish-aurorarp-8x7b-v4/flywheel_model.1.safetensors
fredithefish-aurorarp-8x7b-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/fredithefish-aurorarp-8x7b-v4/flywheel_model.0.safetensors
fredithefish-aurorarp-8x7b-v4-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
fredithefish-aurorarp-8x7b-v4-mkmlizer: 
Loading 0:   0%|          | 0/995 [00:00<?, ?it/s]
Loading 0:   1%|          | 10/995 [00:00<00:10, 90.87it/s]
Loading 0:   2%|▏         | 20/995 [00:00<00:10, 91.16it/s]
Loading 0:   4%|▎         | 36/995 [00:00<00:08, 117.64it/s]
Loading 0:   5%|▍         | 48/995 [00:00<00:08, 107.79it/s]
Loading 0:   6%|▋         | 64/995 [00:00<00:07, 121.81it/s]
Loading 0:   8%|▊         | 77/995 [00:00<00:08, 110.29it/s]
Loading 0:   9%|▉         | 89/995 [00:00<00:08, 109.88it/s]
Loading 0:  11%|█▏        | 112/995 [00:00<00:06, 143.08it/s]
Loading 0:  13%|█▎        | 127/995 [00:01<00:15, 56.31it/s] 
Loading 0:  14%|█▍        | 139/995 [00:01<00:13, 64.67it/s]
Loading 0:  15%|█▌        | 151/995 [00:01<00:11, 70.47it/s]
Loading 0:  17%|█▋        | 166/995 [00:01<00:09, 83.38it/s]
Loading 0:  18%|█▊        | 178/995 [00:02<00:09, 85.44it/s]
Loading 0:  19%|█▉        | 190/995 [00:02<00:08, 91.71it/s]
Loading 0:  20%|██        | 201/995 [00:02<00:08, 92.94it/s]
Loading 0:  22%|██▏       | 216/995 [00:02<00:14, 53.97it/s]
Loading 0:  24%|██▎       | 234/995 [00:02<00:10, 71.94it/s]
Loading 0:  25%|██▍       | 245/995 [00:02<00:09, 78.49it/s]
Loading 0:  26%|██▌       | 256/995 [00:03<00:08, 82.16it/s]
Loading 0:  27%|██▋       | 272/995 [00:03<00:07, 98.56it/s]
Loading 0:  29%|██▊       | 285/995 [00:03<00:07, 98.13it/s]
Loading 0:  30%|██▉       | 297/995 [00:12<02:30,  4.64it/s]
Loading 0:  31%|███       | 307/995 [00:12<01:53,  6.07it/s]
Loading 0:  32%|███▏      | 317/995 [00:12<01:24,  8.04it/s]
Loading 0:  33%|███▎      | 326/995 [00:13<01:11,  9.35it/s]
Loading 0:  34%|███▍      | 336/995 [00:13<00:52, 12.58it/s]
Loading 0:  35%|███▍      | 346/995 [00:13<00:38, 16.79it/s]
Loading 0:  36%|███▌      | 358/995 [00:13<00:27, 23.34it/s]
Loading 0:  37%|███▋      | 368/995 [00:13<00:21, 29.69it/s]
Loading 0:  38%|███▊      | 380/995 [00:13<00:15, 39.08it/s]
Loading 0:  39%|███▉      | 390/995 [00:13<00:12, 46.82it/s]
Loading 0:  41%|████      | 406/995 [00:13<00:09, 64.04it/s]
Loading 0:  42%|████▏     | 418/995 [00:13<00:08, 71.78it/s]
Loading 0:  43%|████▎     | 430/995 [00:14<00:12, 46.24it/s]
Loading 0:  44%|████▍     | 440/995 [00:14<00:10, 53.55it/s]
Loading 0:  45%|████▌     | 452/995 [00:14<00:08, 64.11it/s]
Loading 0:  46%|████▋     | 462/995 [00:14<00:07, 70.13it/s]
Loading 0:  48%|████▊     | 478/995 [00:14<00:05, 87.94it/s]
Loading 0:  49%|████▉     | 490/995 [00:14<00:05, 89.80it/s]
Loading 0:  50%|█████     | 502/995 [00:15<00:05, 95.12it/s]
Loading 0:  52%|█████▏    | 513/995 [00:15<00:05, 95.13it/s]
Loading 0:  53%|█████▎    | 527/995 [00:15<00:04, 105.50it/s]
Loading 0:  54%|█████▍    | 539/995 [00:15<00:08, 51.94it/s] 
Loading 0:  55%|█████▌    | 549/995 [00:15<00:07, 58.55it/s]
Loading 0:  56%|█████▌    | 559/995 [00:16<00:06, 65.27it/s]
Loading 0:  57%|█████▋    | 569/995 [00:25<01:55,  3.70it/s]
Loading 0:  58%|█████▊    | 579/995 [00:25<01:21,  5.09it/s]
Loading 0:  59%|█████▉    | 591/995 [00:25<00:54,  7.41it/s]
Loading 0:  60%|██████    | 601/995 [00:25<00:39, 10.00it/s]
Loading 0:  61%|██████▏   | 611/995 [00:25<00:28, 13.45it/s]
Loading 0:  63%|██████▎   | 627/995 [00:25<00:17, 21.08it/s]
Loading 0:  65%|██████▌   | 650/995 [00:26<00:12, 27.38it/s]
Loading 0:  66%|██████▋   | 660/995 [00:26<00:10, 32.42it/s]
Loading 0:  68%|██████▊   | 672/995 [00:26<00:08, 40.33it/s]
Loading 0:  69%|██████▊   | 682/995 [00:26<00:06, 46.82it/s]
Loading 0:  70%|██████▉   | 692/995 [00:26<00:05, 53.62it/s]
Loading 0:  71%|███████   | 702/995 [00:26<00:04, 60.38it/s]
Loading 0:  72%|███████▏  | 712/995 [00:27<00:04, 67.29it/s]
Loading 0:  73%|███████▎  | 724/995 [00:27<00:03, 77.65it/s]
Loading 0:  74%|███████▍  | 736/995 [00:27<00:02, 87.22it/s]
Loading 0:  75%|███████▌  | 747/995 [00:27<00:05, 46.45it/s]
Loading 0:  76%|███████▌  | 758/995 [00:27<00:04, 55.92it/s]
Loading 0:  77%|███████▋  | 768/995 [00:27<00:03, 62.98it/s]
Loading 0:  79%|███████▊  | 783/995 [00:28<00:02, 79.96it/s]
Loading 0:  80%|███████▉  | 794/995 [00:28<00:02, 85.30it/s]
Loading 0:  81%|████████  | 805/995 [00:28<00:02, 87.65it/s]
Loading 0:  82%|████████▏ | 817/995 [00:28<00:01, 93.77it/s]
Loading 0:  83%|████████▎ | 828/995 [00:28<00:01, 94.42it/s]
Loading 0:  85%|████████▍ | 843/995 [00:37<00:33,  4.50it/s]
Loading 0:  86%|████████▌ | 857/995 [00:39<00:26,  5.29it/s]
Loading 0:  87%|████████▋ | 867/995 [00:39<00:18,  6.91it/s]
Loading 0:  88%|████████▊ | 879/995 [00:39<00:12,  9.60it/s]
Loading 0:  89%|████████▉ | 888/995 [00:39<00:08, 12.25it/s]
Loading 0:  90%|█████████ | 898/995 [00:39<00:06, 16.15it/s]
Loading 0:  91%|█████████▏| 910/995 [00:39<00:03, 22.31it/s]
Loading 0:  92%|█████████▏| 920/995 [00:39<00:02, 28.13it/s]
Loading 0:  94%|█████████▍| 936/995 [00:39<00:01, 41.10it/s]
Loading 0:  95%|█████████▌| 947/995 [00:40<00:00, 48.50it/s]
Loading 0:  97%|█████████▋| 962/995 [00:40<00:00, 41.02it/s]
Loading 0:  98%|█████████▊| 974/995 [00:40<00:00, 50.21it/s]
Loading 0:  99%|█████████▉| 984/995 [00:40<00:00, 57.21it/s]
Loading 0: 100%|█████████▉| 994/995 [00:40<00:00, 63.68it/s]
                                                            
/opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:1096: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
fredithefish-aurorarp-8x7b-v4-mkmlizer:   warnings.warn(
fredithefish-aurorarp-8x7b-v4-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:720: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
fredithefish-aurorarp-8x7b-v4-mkmlizer:   warnings.warn(
fredithefish-aurorarp-8x7b-v4-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:466: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
fredithefish-aurorarp-8x7b-v4-mkmlizer:   warnings.warn(
fredithefish-aurorarp-8x7b-v4-mkmlizer: /opt/conda/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly.  To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
fredithefish-aurorarp-8x7b-v4-mkmlizer:   return self.fget.__get__(instance, owner)()
fredithefish-aurorarp-8x7b-v4-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
fredithefish-aurorarp-8x7b-v4-mkmlizer: Saving duration: 0.224s
fredithefish-aurorarp-8x7b-v4-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.240s
Job fredithefish-aurorarp-8x7b-v4-mkmlizer completed after 158.5s with status: succeeded
Stopping job with name fredithefish-aurorarp-8x7b-v4-mkmlizer
Pipeline stage MKMLizer completed in 163.54s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service fredithefish-aurorarp-8x7b-v4
Waiting for inference service fredithefish-aurorarp-8x7b-v4 to be ready
Inference service fredithefish-aurorarp-8x7b-v4 ready after 70.49562311172485s
Pipeline stage ISVCDeployer completed in 78.40s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.691495895385742s
Received healthy response to inference request in 2.8668720722198486s
Received healthy response to inference request in 2.5026309490203857s
Received healthy response to inference request in 2.558368444442749s
Received healthy response to inference request in 2.7531135082244873s
5 requests
0 failed requests
5th percentile: 2.5137784481048584
10th percentile: 2.524925947189331
20th percentile: 2.5472209453582764
30th percentile: 2.5973174571990967
40th percentile: 2.675215482711792
50th percentile: 2.7531135082244873
60th percentile: 2.7986169338226317
70th percentile: 2.8441203594207765
80th percentile: 3.0317968368530277
90th percentile: 3.361646366119385
95th percentile: 3.526571130752563
99th percentile: 3.6585109424591065
mean time: 2.8744961738586428
Pipeline stage StressChecker completed in 15.01s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.04s
Running pipeline stage DaemonicSafetyScorer
Running M-Eval for topic stay_in_character
Pipeline stage DaemonicSafetyScorer completed in 0.05s
M-Eval Dataset for topic stay_in_character is loaded
fredithefish-aurorarp-8x7b_v4 status is now deployed due to DeploymentManager action
fredithefish-aurorarp-8x7b_v4 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of fredithefish-aurorarp-8x7b_v4
Running pipeline stage ISVCDeleter
Checking if service fredithefish-aurorarp-8x7b-v4 is running
Tearing down inference service fredithefish-aurorarp-8x7b-v4
Toredown service fredithefish-aurorarp-8x7b-v4
Pipeline stage ISVCDeleter completed in 3.98s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key fredithefish-aurorarp-8x7b-v4/config.json from bucket guanaco-mkml-models
Deleting key fredithefish-aurorarp-8x7b-v4/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key fredithefish-aurorarp-8x7b-v4/flywheel_model.1.safetensors from bucket guanaco-mkml-models
Deleting key fredithefish-aurorarp-8x7b-v4/flywheel_model.2.safetensors from bucket guanaco-mkml-models
Deleting key fredithefish-aurorarp-8x7b-v4/flywheel_model.3.safetensors from bucket guanaco-mkml-models
Deleting key fredithefish-aurorarp-8x7b-v4/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key fredithefish-aurorarp-8x7b-v4/tokenizer.json from bucket guanaco-mkml-models
Deleting key fredithefish-aurorarp-8x7b-v4/tokenizer.model from bucket guanaco-mkml-models
Deleting key fredithefish-aurorarp-8x7b-v4/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key fredithefish-aurorarp-8x7b-v4_reward/config.json from bucket guanaco-reward-models
Deleting key fredithefish-aurorarp-8x7b-v4_reward/merges.txt from bucket guanaco-reward-models
Deleting key fredithefish-aurorarp-8x7b-v4_reward/reward.tensors from bucket guanaco-reward-models
Deleting key fredithefish-aurorarp-8x7b-v4_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key fredithefish-aurorarp-8x7b-v4_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key fredithefish-aurorarp-8x7b-v4_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key fredithefish-aurorarp-8x7b-v4_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.67s
fredithefish-aurorarp-8x7b_v4 status is now torndown due to DeploymentManager action