submission_id: jellywibble-meseca-20062_6532_v1
developer_uid: Jellywibble
status: inactive
model_repo: Jellywibble/meseca-20062024-c1
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 60, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
formatter: {'memory_template': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}:', 'truncate_by_message': False}
reward_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}
timestamp: 2024-07-05T19:52:44+00:00
model_name: meseca-20062024-c1_v1
model_group: Jellywibble/meseca-20062
num_battles: 29971
num_wins: 16294
celo_rating: 1207.94
propriety_score: 0.7067757009345794
propriety_total_count: 5136.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: meseca-20062024-c1_v1
ineligible_reason: None
language_model: Jellywibble/meseca-20062024-c1
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-07-05
win_ratio: 0.5436588702412332
Resubmit model
Running pipeline stage MKMLizer
Starting job with name jellywibble-meseca-20062-6532-v1-mkmlizer
Waiting for job on jellywibble-meseca-20062-6532-v1-mkmlizer to finish
jellywibble-meseca-20062-6532-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ _____ __ __ ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ /___/ ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ Version: 0.8.14 ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ https://mk1.ai ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ The license key for the current software has been verified as ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ belonging to: ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ Chai Research Corp. ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ║ ║
jellywibble-meseca-20062-6532-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Failed to get response for submission mistralai-mixtral-8x7b-_3473_v61: ('http://mistralai-mixtral-8x7b-3473-v61-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'upstream connect error or disconnect/reset before headers. reset reason: connection timeout')
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
jellywibble-meseca-20062-6532-v1-mkmlizer: Downloaded to shared memory in 66.319s
jellywibble-meseca-20062-6532-v1-mkmlizer: quantizing model to /dev/shm/model_cache
jellywibble-meseca-20062-6532-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jellywibble-meseca-20062-6532-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 3/291 [00:00<00:56, 5.10it/s] Loading 0: 1%|▏ | 4/291 [00:01<01:27, 3.28it/s] Loading 0: 2%|▏ | 5/291 [00:01<01:33, 3.05it/s] Loading 0: 3%|▎ | 8/291 [00:01<00:49, 5.71it/s] Loading 0: 3%|▎ | 9/291 [00:01<00:49, 5.74it/s] Loading 0: 4%|▍ | 12/291 [00:02<00:51, 5.38it/s] Loading 0: 4%|▍ | 13/291 [00:02<01:04, 4.30it/s] Loading 0: 5%|▍ | 14/291 [00:03<01:16, 3.60it/s] Loading 0: 6%|▌ | 17/291 [00:03<00:46, 5.87it/s] Loading 0: 6%|▌ | 18/291 [00:03<00:43, 6.24it/s] Loading 0: 7%|▋ | 21/291 [00:04<00:40, 6.59it/s] Loading 0: 8%|▊ | 22/291 [00:04<00:53, 5.05it/s] Loading 0: 8%|▊ | 23/291 [00:05<01:12, 3.70it/s] Loading 0: 9%|▉ | 26/291 [00:05<00:46, 5.66it/s] Loading 0: 9%|▉ | 27/291 [00:05<00:45, 5.78it/s] Loading 0: 10%|█ | 30/291 [00:05<00:34, 7.61it/s] Loading 0: 11%|█ | 31/291 [00:05<00:34, 7.43it/s] Loading 0: 11%|█▏ | 33/291 [00:06<00:37, 6.96it/s] Loading 0: 12%|█▏ | 34/291 [00:06<00:58, 4.40it/s] Loading 0: 12%|█▏ | 35/291 [00:07<01:13, 3.50it/s] Loading 0: 12%|█▏ | 36/291 [00:07<01:26, 2.95it/s] Loading 0: 13%|█▎ | 39/291 [00:08<01:05, 3.85it/s] Loading 0: 14%|█▎ | 40/291 [00:08<01:18, 3.18it/s] Loading 0: 14%|█▍ | 41/291 [00:09<01:33, 2.67it/s] Loading 0: 15%|█▌ | 44/291 [00:09<00:54, 4.51it/s] Loading 0: 15%|█▌ | 45/291 [00:09<00:50, 4.91it/s] Loading 0: 16%|█▋ | 48/291 [00:10<00:48, 5.05it/s] Loading 0: 17%|█▋ | 49/291 [00:10<01:00, 4.01it/s] Loading 0: 17%|█▋ | 50/291 [00:11<01:10, 3.40it/s] Loading 0: 18%|█▊ | 53/291 [00:11<00:44, 5.37it/s] Loading 0: 19%|█▊ | 54/291 [00:11<00:41, 5.78it/s] Loading 0: 20%|█▉ | 57/291 [00:11<00:40, 5.81it/s] Loading 0: 20%|█▉ | 58/291 [00:12<00:52, 4.45it/s] Loading 0: 20%|██ | 59/291 [00:12<00:59, 3.88it/s] Loading 0: 21%|██▏ | 62/291 [00:12<00:36, 6.24it/s] Loading 0: 22%|██▏ | 64/291 [00:13<00:30, 7.52it/s] Loading 0: 23%|██▎ | 66/291 [00:13<00:25, 8.83it/s] Loading 0: 24%|██▍ | 71/291 [00:13<00:14, 15.35it/s] Loading 0: 26%|██▌ | 75/291 [00:13<00:10, 19.70it/s] Loading 0: 27%|██▋ | 80/291 [00:13<00:11, 18.24it/s] Loading 0: 29%|██▊ | 83/291 [00:13<00:10, 19.30it/s] Loading 0: 31%|███▏ | 91/291 [00:14<00:06, 29.89it/s] Loading 0: 34%|███▍ | 99/291 [00:14<00:04, 39.05it/s] Loading 0: 36%|███▌ | 104/291 [00:14<00:04, 41.14it/s] Loading 0: 38%|███▊ | 110/291 [00:14<00:04, 41.56it/s] Loading 0: 41%|████ | 119/291 [00:14<00:03, 53.09it/s] Loading 0: 43%|████▎ | 126/291 [00:14<00:03, 53.48it/s] Loading 0: 46%|████▌ | 133/291 [00:14<00:04, 36.86it/s] Loading 0: 47%|████▋ | 138/291 [00:15<00:04, 37.16it/s] Loading 0: 49%|████▉ | 143/291 [00:15<00:03, 39.53it/s] Loading 0: 51%|█████ | 148/291 [00:15<00:03, 41.70it/s] Loading 0: 54%|█████▎ | 156/291 [00:15<00:02, 49.53it/s] Loading 0: 56%|█████▌ | 162/291 [00:15<00:02, 51.06it/s] Loading 0: 58%|█████▊ | 168/291 [00:15<00:02, 49.66it/s] Loading 0: 60%|██████ | 175/291 [00:15<00:02, 53.60it/s] Loading 0: 64%|██████▍ | 186/291 [00:16<00:02, 41.26it/s] Loading 0: 66%|██████▌ | 191/291 [00:16<00:02, 41.44it/s] Loading 0: 67%|██████▋ | 196/291 [00:16<00:02, 41.98it/s] Loading 0: 70%|██████▉ | 203/291 [00:16<00:01, 44.75it/s] Loading 0: 73%|███████▎ | 211/291 [00:16<00:01, 51.82it/s] Loading 0: 75%|███████▌ | 219/291 [00:16<00:01, 57.99it/s] Loading 0: 78%|███████▊ | 226/291 [00:16<00:01, 59.59it/s] Loading 0: 80%|████████ | 233/291 [00:17<00:01, 37.22it/s] Loading 0: 82%|████████▏ | 238/291 [00:17<00:01, 38.14it/s] Loading 0: 85%|████████▍ | 246/291 [00:17<00:00, 45.38it/s] Loading 0: 87%|████████▋ | 252/291 [00:17<00:00, 47.65it/s] Loading 0: 89%|████████▊ | 258/291 [00:17<00:00, 47.51it/s] Loading 0: 91%|█████████ | 265/291 [00:17<00:00, 51.76it/s] Loading 0: 94%|█████████▍| 273/291 [00:17<00:00, 56.80it/s] Loading 0: 96%|█████████▌| 280/291 [00:17<00:00, 58.55it/s] Loading 0: 99%|█████████▊| 287/291 [00:23<00:01, 3.87it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
jellywibble-meseca-20062-6532-v1-mkmlizer: quantized model in 73.826s
jellywibble-meseca-20062-6532-v1-mkmlizer: Processed model Jellywibble/meseca-20062024-c1 in 140.146s
jellywibble-meseca-20062-6532-v1-mkmlizer: creating bucket guanaco-mkml-models
jellywibble-meseca-20062-6532-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jellywibble-meseca-20062-6532-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jellywibble-meseca-20062-6532-v1
jellywibble-meseca-20062-6532-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jellywibble-meseca-20062-6532-v1/config.json
jellywibble-meseca-20062-6532-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jellywibble-meseca-20062-6532-v1/tokenizer_config.json
jellywibble-meseca-20062-6532-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jellywibble-meseca-20062-6532-v1/special_tokens_map.json
jellywibble-meseca-20062-6532-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jellywibble-meseca-20062-6532-v1/tokenizer.json
jellywibble-meseca-20062-6532-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jellywibble-meseca-20062-6532-v1/flywheel_model.0.safetensors
jellywibble-meseca-20062-6532-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
jellywibble-meseca-20062-6532-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:919: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jellywibble-meseca-20062-6532-v1-mkmlizer: warnings.warn(
jellywibble-meseca-20062-6532-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
jellywibble-meseca-20062-6532-v1-mkmlizer: warnings.warn(
jellywibble-meseca-20062-6532-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:769: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jellywibble-meseca-20062-6532-v1-mkmlizer: warnings.warn(
jellywibble-meseca-20062-6532-v1-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.
jellywibble-meseca-20062-6532-v1-mkmlizer: warnings.warn(
jellywibble-meseca-20062-6532-v1-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()
jellywibble-meseca-20062-6532-v1-mkmlizer: return self.fget.__get__(instance, owner)()
jellywibble-meseca-20062-6532-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jellywibble-meseca-20062-6532-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jellywibble-meseca-20062-6532-v1_reward/tokenizer.json
jellywibble-meseca-20062-6532-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jellywibble-meseca-20062-6532-v1_reward/reward.tensors
Job jellywibble-meseca-20062-6532-v1-mkmlizer completed after 165.38s with status: succeeded
Stopping job with name jellywibble-meseca-20062-6532-v1-mkmlizer
Pipeline stage MKMLizer completed in 166.29s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.12s
Running pipeline stage ISVCDeployer
Creating inference service jellywibble-meseca-20062-6532-v1
Waiting for inference service jellywibble-meseca-20062-6532-v1 to be ready
Inference service jellywibble-meseca-20062-6532-v1 ready after 40.29145550727844s
Pipeline stage ISVCDeployer completed in 47.41s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.074164867401123s
Received healthy response to inference request in 1.3113188743591309s
Received healthy response to inference request in 1.2966067790985107s
Received healthy response to inference request in 1.2846364974975586s
Received healthy response to inference request in 1.3414523601531982s
5 requests
0 failed requests
5th percentile: 1.2870305538177491
10th percentile: 1.2894246101379394
20th percentile: 1.2942127227783202
30th percentile: 1.2995491981506349
40th percentile: 1.3054340362548829
50th percentile: 1.3113188743591309
60th percentile: 1.3233722686767577
70th percentile: 1.3354256629943848
80th percentile: 1.4879948616027834
90th percentile: 1.7810798645019532
95th percentile: 1.927622365951538
99th percentile: 2.044856367111206
mean time: 1.4616358757019043
Pipeline stage StressChecker completed in 8.05s
jellywibble-meseca-20062_6532_v1 status is now deployed due to DeploymentManager action
jellywibble-meseca-20062_6532_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics