r136a1-ayam-2x8b

developer_uid: R136a1

submission_id: r136a1-ayam-2x8b_v3

model_name: r136a1-ayam-2x8b_v1

model_group: R136a1/Ayam-2x8B

status: torndown

timestamp: 2024-06-11T21:25:57+00:00

num_battles: 25149

num_wins: 13376

celo_rating: 1207.01

family_friendly_score: 0.0

submission_type: basic

model_repo: R136a1/Ayam-2x8B

model_architecture: MixtralForCausalLM

reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2

model_num_parameters: 13667667968.0

best_of: 8

max_input_tokens: 512

max_output_tokens: 64

display_name: r136a1-ayam-2x8b_v1

is_internal_developer: False

language_model: R136a1/Ayam-2x8B

model_size: 14B

ranking_group: single

us_pacific_date: 2024-06-11

win_ratio: 0.5318700544753271

generation_params: {'temperature': 1.15, 'top_p': 1.0, 'min_p': 0.075, 'top_k': 70, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n{bot_name}:', 'truncate_by_message': False}

model_eval_status: success

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 r136a1-ayam-2x8b-v3-mkmlizer
Waiting for job on r136a1-ayam-2x8b-v3-mkmlizer to finish
r136a1-ayam-2x8b-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
r136a1-ayam-2x8b-v3-mkmlizer: ║     _____            __           __                                ║
r136a1-ayam-2x8b-v3-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
r136a1-ayam-2x8b-v3-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
r136a1-ayam-2x8b-v3-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
r136a1-ayam-2x8b-v3-mkmlizer: ║       /___/                                                         ║
r136a1-ayam-2x8b-v3-mkmlizer: ║                                                                     ║
r136a1-ayam-2x8b-v3-mkmlizer: ║  Version: 0.8.14                                                    ║
r136a1-ayam-2x8b-v3-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
r136a1-ayam-2x8b-v3-mkmlizer: ║  https://mk1.ai                                                     ║
r136a1-ayam-2x8b-v3-mkmlizer: ║                                                                     ║
r136a1-ayam-2x8b-v3-mkmlizer: ║  The license key for the current software has been verified as      ║
r136a1-ayam-2x8b-v3-mkmlizer: ║  belonging to:                                                      ║
r136a1-ayam-2x8b-v3-mkmlizer: ║                                                                     ║
r136a1-ayam-2x8b-v3-mkmlizer: ║  Chai Research Corp.                                                ║
r136a1-ayam-2x8b-v3-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
r136a1-ayam-2x8b-v3-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
r136a1-ayam-2x8b-v3-mkmlizer: ║                                                                     ║
r136a1-ayam-2x8b-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
r136a1-ayam-2x8b-v3-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.
r136a1-ayam-2x8b-v3-mkmlizer:   warnings.warn(warning_message, FutureWarning)
r136a1-ayam-2x8b-v3-mkmlizer: Downloaded to shared memory in 22.172s
r136a1-ayam-2x8b-v3-mkmlizer: quantizing model to /dev/shm/model_cache
r136a1-ayam-2x8b-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
r136a1-ayam-2x8b-v3-mkmlizer: 
Loading 0:   0%|          | 0/419 [00:00<?, ?it/s]
Loading 0:   4%|▍         | 16/419 [00:00<00:02, 144.49it/s]
Loading 0:   7%|▋         | 31/419 [00:00<00:02, 139.20it/s]
Loading 0:  12%|█▏        | 51/419 [00:00<00:02, 160.88it/s]
Loading 0:  16%|█▌        | 68/419 [00:00<00:04, 80.04it/s] 
Loading 0:  21%|██▏       | 90/419 [00:00<00:03, 108.37it/s]
Loading 0:  25%|██▌       | 106/419 [00:00<00:02, 118.85it/s]
Loading 0:  29%|██▉       | 122/419 [00:01<00:02, 121.99it/s]
Loading 0:  34%|███▎      | 141/419 [00:01<00:03, 86.78it/s] 
Loading 0:  37%|███▋      | 153/419 [00:01<00:02, 90.09it/s]
Loading 0:  41%|████      | 171/419 [00:01<00:02, 106.67it/s]
Loading 0:  44%|████▍     | 186/419 [00:01<00:02, 113.35it/s]
Loading 0:  49%|████▉     | 206/419 [00:01<00:01, 131.25it/s]
Loading 0:  53%|█████▎    | 221/419 [00:02<00:02, 82.44it/s] 
Loading 0:  56%|█████▌    | 235/419 [00:02<00:01, 92.52it/s]
Loading 0:  61%|██████    | 255/419 [00:02<00:01, 112.25it/s]
Loading 0:  64%|██████▍   | 270/419 [00:02<00:01, 117.62it/s]
Loading 0:  69%|██████▉   | 289/419 [00:02<00:01, 85.04it/s] 
Loading 0:  73%|███████▎  | 304/419 [00:02<00:01, 94.35it/s]
Loading 0:  76%|███████▌  | 318/419 [00:03<00:00, 102.10it/s]
Loading 0:  80%|████████  | 336/419 [00:03<00:00, 118.10it/s]
Loading 0:  84%|████████▎ | 350/419 [00:03<00:00, 113.77it/s]
Loading 0:  87%|████████▋ | 365/419 [00:03<00:00, 121.13it/s]
Loading 0:  88%|████████▊ | 367/419 [00:13<00:00, 121.13it/s]
Loading 0:  88%|████████▊ | 368/419 [00:21<00:24,  2.08it/s] 
Loading 0:  95%|█████████▍| 398/419 [00:22<00:04,  4.32it/s]
Loading 0:  99%|█████████▉| 416/419 [00:22<00:00,  6.17it/s]
                                                            
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
r136a1-ayam-2x8b-v3-mkmlizer: quantized model in 28.820s
r136a1-ayam-2x8b-v3-mkmlizer: Processed model R136a1/Ayam-2x8B in 52.668s
r136a1-ayam-2x8b-v3-mkmlizer: creating bucket guanaco-mkml-models
r136a1-ayam-2x8b-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
r136a1-ayam-2x8b-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/r136a1-ayam-2x8b-v3
r136a1-ayam-2x8b-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/r136a1-ayam-2x8b-v3/config.json
r136a1-ayam-2x8b-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/r136a1-ayam-2x8b-v3/special_tokens_map.json
r136a1-ayam-2x8b-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/r136a1-ayam-2x8b-v3/tokenizer_config.json
r136a1-ayam-2x8b-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/r136a1-ayam-2x8b-v3/tokenizer.json
r136a1-ayam-2x8b-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/r136a1-ayam-2x8b-v3/flywheel_model.1.safetensors
r136a1-ayam-2x8b-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/r136a1-ayam-2x8b-v3/flywheel_model.0.safetensors
r136a1-ayam-2x8b-v3-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
r136a1-ayam-2x8b-v3-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:913: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
r136a1-ayam-2x8b-v3-mkmlizer:   warnings.warn(
r136a1-ayam-2x8b-v3-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:757: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
r136a1-ayam-2x8b-v3-mkmlizer:   warnings.warn(
r136a1-ayam-2x8b-v3-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:468: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
r136a1-ayam-2x8b-v3-mkmlizer:   warnings.warn(
r136a1-ayam-2x8b-v3-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()
r136a1-ayam-2x8b-v3-mkmlizer:   return self.fget.__get__(instance, owner)()
r136a1-ayam-2x8b-v3-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
r136a1-ayam-2x8b-v3-mkmlizer: Saving duration: 0.211s
r136a1-ayam-2x8b-v3-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.301s
r136a1-ayam-2x8b-v3-mkmlizer: creating bucket guanaco-reward-models
r136a1-ayam-2x8b-v3-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/r136a1-ayam-2x8b-v3_reward/reward.tensors
Job r136a1-ayam-2x8b-v3-mkmlizer completed after 83.28s with status: succeeded
Stopping job with name r136a1-ayam-2x8b-v3-mkmlizer
Pipeline stage MKMLizer completed in 88.20s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service r136a1-ayam-2x8b-v3
Waiting for inference service r136a1-ayam-2x8b-v3 to be ready
Inference service r136a1-ayam-2x8b-v3 ready after 40.26802110671997s
Pipeline stage ISVCDeployer completed in 48.31s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.7297298908233643s
Received healthy response to inference request in 2.094249963760376s
Received healthy response to inference request in 2.0906789302825928s
Received healthy response to inference request in 2.045262336730957s
Received healthy response to inference request in 2.0911412239074707s
5 requests
0 failed requests
5th percentile: 2.054345655441284
10th percentile: 2.0634289741516114
20th percentile: 2.081595611572266
30th percentile: 2.0907713890075685
40th percentile: 2.0909563064575196
50th percentile: 2.0911412239074707
60th percentile: 2.0923847198486327
70th percentile: 2.0936282157897947
80th percentile: 2.221345949172974
90th percentile: 2.475537919998169
95th percentile: 2.6026339054107663
99th percentile: 2.7043106937408448
mean time: 2.210212469100952
Pipeline stage StressChecker completed in 11.64s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Running M-Eval for topic stay_in_character
Pipeline stage DaemonicSafetyScorer completed in 0.03s
M-Eval Dataset for topic stay_in_character is loaded
r136a1-ayam-2x8b_v3 status is now deployed due to DeploymentManager action
r136a1-ayam-2x8b_v3 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of r136a1-ayam-2x8b_v3
Running pipeline stage ISVCDeleter
Checking if service r136a1-ayam-2x8b-v3 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 4.04s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key r136a1-ayam-2x8b-v3/config.json from bucket guanaco-mkml-models
Deleting key r136a1-ayam-2x8b-v3/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key r136a1-ayam-2x8b-v3/flywheel_model.1.safetensors from bucket guanaco-mkml-models
Deleting key r136a1-ayam-2x8b-v3/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key r136a1-ayam-2x8b-v3/tokenizer.json from bucket guanaco-mkml-models
Deleting key r136a1-ayam-2x8b-v3/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key r136a1-ayam-2x8b-v3_reward/config.json from bucket guanaco-reward-models
Deleting key r136a1-ayam-2x8b-v3_reward/merges.txt from bucket guanaco-reward-models
Deleting key r136a1-ayam-2x8b-v3_reward/reward.tensors from bucket guanaco-reward-models
Deleting key r136a1-ayam-2x8b-v3_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key r136a1-ayam-2x8b-v3_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key r136a1-ayam-2x8b-v3_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key r136a1-ayam-2x8b-v3_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 7.70s
r136a1-ayam-2x8b_v3 status is now torndown due to DeploymentManager action