submission_id: neversleep-noromaid-v0-_8068_v94
developer_uid: robert_irvine
status: inactive
model_repo: NeverSleep/Noromaid-v0.1-mixtral-8x7b-Instruct-v3
reward_repo: rirv938/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 50, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['</s>', '<|user|>', '###', '\n'], 'max_input_tokens': 512, 'best_of': 4, 'max_output_tokens': 64}
formatter: {'memory_template': '<s>[INST] This is an entertaining conversation. You are {bot_name} who has the persona: {memory}.\nPlay the role of {bot_name}. Engage in a chat with {user_name} while staying in character. You should create a fun dialogue which entertains {user_name}.\n', 'prompt_template': '{prompt}\n', 'bot_template': '{bot_name}: {message}</s>', 'user_template': '[INST] {user_name}: {message} [/INST]', 'response_template': '[INST] respond with a plot twist [/INST]{bot_name}:', 'truncate_by_message': False}
reward_formatter: {'memory_template': 'Memory: {memory}\n', 'prompt_template': '{prompt}\n', 'bot_template': 'Bot: {message}\n', 'user_template': 'User: {message}\n', 'response_template': 'Bot:', 'truncate_by_message': False}
timestamp: 2024-07-01T22:43:22+00:00
model_name: neversleep-noromaid-v0-_8068_v94
model_group: NeverSleep/Noromaid-v0.1
num_battles: 13814
num_wins: 6633
celo_rating: 1142.02
propriety_score: 0.7290242386575513
propriety_total_count: 6436.0
submission_type: basic
model_architecture: MixtralForCausalLM
model_num_parameters: 46702792704.0
best_of: 4
max_input_tokens: 512
max_output_tokens: 64
display_name: neversleep-noromaid-v0-_8068_v94
ineligible_reason: None
language_model: NeverSleep/Noromaid-v0.1-mixtral-8x7b-Instruct-v3
model_size: 47B
reward_model: rirv938/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-07-01
win_ratio: 0.4801650499493268
Resubmit model
Running pipeline stage MKMLizer
Starting job with name neversleep-noromaid-v0-8068-v94-mkmlizer
Waiting for job on neversleep-noromaid-v0-8068-v94-mkmlizer to finish
neversleep-noromaid-v0-8068-v94-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ _____ __ __ ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ /___/ ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ Version: 0.8.14 ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ https://mk1.ai ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ The license key for the current software has been verified as ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ belonging to: ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ Chai Research Corp. ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ║ ║
neversleep-noromaid-v0-8068-v94-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
neversleep-noromaid-v0-8068-v94-mkmlizer: Downloaded to shared memory in 113.542s
neversleep-noromaid-v0-8068-v94-mkmlizer: quantizing model to /dev/shm/model_cache
neversleep-noromaid-v0-8068-v94-mkmlizer: Saving flywheel model at /dev/shm/model_cache
neversleep-noromaid-v0-8068-v94-mkmlizer: quantized model in 125.383s
neversleep-noromaid-v0-8068-v94-mkmlizer: Processed model NeverSleep/Noromaid-v0.1-mixtral-8x7b-Instruct-v3 in 238.925s
neversleep-noromaid-v0-8068-v94-mkmlizer: creating bucket guanaco-mkml-models
neversleep-noromaid-v0-8068-v94-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
neversleep-noromaid-v0-8068-v94-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/neversleep-noromaid-v0-8068-v94
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/neversleep-noromaid-v0-8068-v94/tokenizer.model
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/neversleep-noromaid-v0-8068-v94/tokenizer.json
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /dev/shm/model_cache/flywheel_model.3.safetensors s3://guanaco-mkml-models/neversleep-noromaid-v0-8068-v94/flywheel_model.3.safetensors
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/neversleep-noromaid-v0-8068-v94/flywheel_model.0.safetensors
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /dev/shm/model_cache/flywheel_model.2.safetensors s3://guanaco-mkml-models/neversleep-noromaid-v0-8068-v94/flywheel_model.2.safetensors
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/neversleep-noromaid-v0-8068-v94/flywheel_model.1.safetensors
neversleep-noromaid-v0-8068-v94-mkmlizer: loading reward model from rirv938/reward_gpt2_medium_preference_24m_e2
neversleep-noromaid-v0-8068-v94-mkmlizer: Loading 0: 0%| | 0/995 [00:00<?, ?it/s] Loading 0: 5%|▌ | 52/995 [00:08<02:33, 6.15it/s] Loading 0: 11%|█ | 107/995 [00:11<01:27, 10.14it/s] Loading 0: 16%|█▋ | 162/995 [00:19<01:37, 8.52it/s] Loading 0: 21%|██ | 210/995 [00:20<01:04, 12.24it/s] Loading 0: 27%|██▋ | 265/995 [00:21<00:41, 17.44it/s] Loading 0: 28%|██▊ | 277/995 [00:33<00:41, 17.44it/s] Loading 0: 28%|██▊ | 278/995 [00:44<03:01, 3.96it/s] Loading 0: 32%|███▏ | 320/995 [00:45<01:58, 5.69it/s] Loading 0: 37%|███▋ | 368/995 [00:46<01:15, 8.34it/s] Loading 0: 43%|████▎ | 423/995 [00:47<00:46, 12.22it/s] Loading 0: 48%|████▊ | 478/995 [00:48<00:30, 16.84it/s] Loading 0: 53%|█████▎ | 526/995 [00:49<00:22, 21.18it/s] Loading 0: 57%|█████▋ | 564/995 [01:04<00:20, 21.18it/s] Loading 0: 57%|█████▋ | 565/995 [01:13<01:25, 5.04it/s] Loading 0: 58%|█████▊ | 581/995 [01:15<01:15, 5.50it/s] Loading 0: 64%|██████▍ | 636/995 [01:16<00:42, 8.53it/s] Loading 0: 69%|██████▉ | 691/995 [01:16<00:24, 12.33it/s] Loading 0: 74%|███████▍ | 739/995 [01:17<00:15, 16.14it/s] Loading 0: 80%|███████▉ | 794/995 [01:18<00:09, 21.28it/s] Loading 0: 85%|████████▍ | 845/995 [01:34<00:07, 21.28it/s] Loading 0: 85%|████████▌ | 846/995 [01:41<00:25, 5.90it/s] Loading 0: 85%|████████▌ | 849/995 [01:42<00:25, 5.70it/s] Loading 0: 90%|█████████ | 897/995 [01:44<00:11, 8.26it/s] Loading 0: 96%|█████████▌| 952/995 [01:45<00:03, 12.12it/s] Loading 0: 100%|██████████| 995/995 [01:46<00:00, 14.37it/s] /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.
neversleep-noromaid-v0-8068-v94-mkmlizer: warnings.warn(
neversleep-noromaid-v0-8068-v94-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`.
neversleep-noromaid-v0-8068-v94-mkmlizer: warnings.warn(
neversleep-noromaid-v0-8068-v94-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.
neversleep-noromaid-v0-8068-v94-mkmlizer: warnings.warn(
neversleep-noromaid-v0-8068-v94-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.
neversleep-noromaid-v0-8068-v94-mkmlizer: warnings.warn(
neversleep-noromaid-v0-8068-v94-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()
neversleep-noromaid-v0-8068-v94-mkmlizer: return self.fget.__get__(instance, owner)()
neversleep-noromaid-v0-8068-v94-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
neversleep-noromaid-v0-8068-v94-mkmlizer: Saving duration: 0.389s
neversleep-noromaid-v0-8068-v94-mkmlizer: Processed model rirv938/reward_gpt2_medium_preference_24m_e2 in 4.941s
neversleep-noromaid-v0-8068-v94-mkmlizer: creating bucket guanaco-reward-models
neversleep-noromaid-v0-8068-v94-mkmlizer: Bucket 's3://guanaco-reward-models/' created
neversleep-noromaid-v0-8068-v94-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/neversleep-noromaid-v0-8068-v94_reward
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/neversleep-noromaid-v0-8068-v94_reward/config.json
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/neversleep-noromaid-v0-8068-v94_reward/special_tokens_map.json
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/neversleep-noromaid-v0-8068-v94_reward/tokenizer_config.json
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/neversleep-noromaid-v0-8068-v94_reward/merges.txt
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/neversleep-noromaid-v0-8068-v94_reward/vocab.json
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/neversleep-noromaid-v0-8068-v94_reward/tokenizer.json
neversleep-noromaid-v0-8068-v94-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/neversleep-noromaid-v0-8068-v94_reward/reward.tensors
Job neversleep-noromaid-v0-8068-v94-mkmlizer completed after 332.64s with status: succeeded
Stopping job with name neversleep-noromaid-v0-8068-v94-mkmlizer
Pipeline stage MKMLizer completed in 333.54s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service neversleep-noromaid-v0-8068-v94
Waiting for inference service neversleep-noromaid-v0-8068-v94 to be ready
Connection pool is full, discarding connection: %s
Inference service neversleep-noromaid-v0-8068-v94 ready after 150.7180211544037s
Pipeline stage ISVCDeployer completed in 157.49s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.0346570014953613s
Received healthy response to inference request in 2.2848403453826904s
Received healthy response to inference request in 2.2670366764068604s
Received healthy response to inference request in 2.228034734725952s
Received healthy response to inference request in 2.2226195335388184s
5 requests
0 failed requests
5th percentile: 2.223702573776245
10th percentile: 2.2247856140136717
20th percentile: 2.2269516944885255
30th percentile: 2.2358351230621336
40th percentile: 2.251435899734497
50th percentile: 2.2670366764068604
60th percentile: 2.2741581439971923
70th percentile: 2.2812796115875242
80th percentile: 2.434803676605225
90th percentile: 2.734730339050293
95th percentile: 2.884693670272827
99th percentile: 3.0046643352508546
mean time: 2.4074376583099366
Pipeline stage StressChecker completed in 12.80s
neversleep-noromaid-v0-_8068_v94 status is now deployed due to DeploymentManager action
neversleep-noromaid-v0-_8068_v94 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics