submission_id: grimjim-llama-3-nephilim_2191_v1
developer_uid: grimjim.
status: inactive
model_repo: grimjim/llama-3-Nephilim-v1-8B
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.2, 'top_p': 1.0, 'min_p': 0.01, 'top_k': 200, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
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}
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: 2024-06-21T14:38:02+00:00
model_name: grimjim-llama-3-nephilim_2191_v1
model_group: grimjim/llama-3-Nephilim
num_battles: 23593
num_wins: 11460
celo_rating: 1180.09
propriety_score: 0.7225140211875724
propriety_total_count: 11233.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: grimjim-llama-3-nephilim_2191_v1
ineligible_reason: None
language_model: grimjim/llama-3-Nephilim-v1-8B
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-21
win_ratio: 0.4857372949603696
Resubmit model
Running pipeline stage MKMLizer
Starting job with name grimjim-llama-3-nephilim-2191-v1-mkmlizer
Waiting for job on grimjim-llama-3-nephilim-2191-v1-mkmlizer to finish
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ _____ __ __ ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ /___/ ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ Version: 0.8.14 ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ https://mk1.ai ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ The license key for the current software has been verified as ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ belonging to: ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ Chai Research Corp. ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ║ ║
grimjim-llama-3-nephilim-2191-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
grimjim-llama-3-nephilim-2191-v1-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.
grimjim-llama-3-nephilim-2191-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
grimjim-llama-3-nephilim-2191-v1-mkmlizer: Downloaded to shared memory in 31.871s
grimjim-llama-3-nephilim-2191-v1-mkmlizer: quantizing model to /dev/shm/model_cache
grimjim-llama-3-nephilim-2191-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
grimjim-llama-3-nephilim-2191-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<10:45, 2.23s/it] Loading 0: 6%|▌ | 17/291 [00:04<00:53, 5.08it/s] Loading 0: 11%|█▏ | 33/291 [00:04<00:22, 11.62it/s] Loading 0: 18%|█▊ | 51/291 [00:04<00:11, 21.13it/s] Loading 0: 22%|██▏ | 65/291 [00:05<00:08, 25.36it/s] Loading 0: 27%|██▋ | 79/291 [00:05<00:06, 34.86it/s] Loading 0: 33%|███▎ | 96/291 [00:05<00:04, 48.74it/s] Loading 0: 39%|███▉ | 114/291 [00:05<00:02, 64.92it/s] Loading 0: 45%|████▌ | 132/291 [00:05<00:01, 81.86it/s] Loading 0: 52%|█████▏ | 150/291 [00:05<00:01, 97.48it/s] Loading 0: 57%|█████▋ | 166/291 [00:05<00:01, 75.61it/s] Loading 0: 63%|██████▎ | 184/291 [00:06<00:01, 91.17it/s] Loading 0: 69%|██████▉ | 202/291 [00:06<00:00, 106.67it/s] Loading 0: 75%|███████▍ | 217/291 [00:06<00:00, 112.52it/s] Loading 0: 80%|███████▉ | 232/291 [00:06<00:00, 120.09it/s] Loading 0: 86%|████████▌ | 249/291 [00:06<00:00, 128.91it/s] Loading 0: 91%|█████████▏| 266/291 [00:06<00:00, 86.61it/s] Loading 0: 97%|█████████▋| 283/291 [00:06<00:00, 99.12it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
grimjim-llama-3-nephilim-2191-v1-mkmlizer: quantized model in 23.091s
grimjim-llama-3-nephilim-2191-v1-mkmlizer: Processed model grimjim/llama-3-Nephilim-v1-8B in 57.505s
grimjim-llama-3-nephilim-2191-v1-mkmlizer: creating bucket guanaco-mkml-models
grimjim-llama-3-nephilim-2191-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
grimjim-llama-3-nephilim-2191-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/grimjim-llama-3-nephilim-2191-v1
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/grimjim-llama-3-nephilim-2191-v1/config.json
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/grimjim-llama-3-nephilim-2191-v1/special_tokens_map.json
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/grimjim-llama-3-nephilim-2191-v1/tokenizer_config.json
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/grimjim-llama-3-nephilim-2191-v1/tokenizer.json
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/grimjim-llama-3-nephilim-2191-v1/flywheel_model.0.safetensors
grimjim-llama-3-nephilim-2191-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
grimjim-llama-3-nephilim-2191-v1-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.
grimjim-llama-3-nephilim-2191-v1-mkmlizer: warnings.warn(
grimjim-llama-3-nephilim-2191-v1-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.
grimjim-llama-3-nephilim-2191-v1-mkmlizer: warnings.warn(
grimjim-llama-3-nephilim-2191-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.
grimjim-llama-3-nephilim-2191-v1-mkmlizer: warnings.warn(
grimjim-llama-3-nephilim-2191-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()
grimjim-llama-3-nephilim-2191-v1-mkmlizer: return self.fget.__get__(instance, owner)()
grimjim-llama-3-nephilim-2191-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
grimjim-llama-3-nephilim-2191-v1-mkmlizer: Saving duration: 0.407s
grimjim-llama-3-nephilim-2191-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.275s
grimjim-llama-3-nephilim-2191-v1-mkmlizer: creating bucket guanaco-reward-models
grimjim-llama-3-nephilim-2191-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
grimjim-llama-3-nephilim-2191-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/grimjim-llama-3-nephilim-2191-v1_reward
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/grimjim-llama-3-nephilim-2191-v1_reward/config.json
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/grimjim-llama-3-nephilim-2191-v1_reward/special_tokens_map.json
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/grimjim-llama-3-nephilim-2191-v1_reward/tokenizer_config.json
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/grimjim-llama-3-nephilim-2191-v1_reward/merges.txt
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/grimjim-llama-3-nephilim-2191-v1_reward/vocab.json
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/grimjim-llama-3-nephilim-2191-v1_reward/tokenizer.json
grimjim-llama-3-nephilim-2191-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/grimjim-llama-3-nephilim-2191-v1_reward/reward.tensors
Job grimjim-llama-3-nephilim-2191-v1-mkmlizer completed after 94.06s with status: succeeded
Stopping job with name grimjim-llama-3-nephilim-2191-v1-mkmlizer
Pipeline stage MKMLizer completed in 94.52s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.13s
Running pipeline stage ISVCDeployer
Creating inference service grimjim-llama-3-nephilim-2191-v1
Waiting for inference service grimjim-llama-3-nephilim-2191-v1 to be ready
Inference service grimjim-llama-3-nephilim-2191-v1 ready after 40.21607685089111s
Pipeline stage ISVCDeployer completed in 45.97s
Running pipeline stage StressChecker
Received healthy response to inference request in 4.971768856048584s
Received healthy response to inference request in 1.374140739440918s
Received healthy response to inference request in 1.2733955383300781s
Received healthy response to inference request in 1.2721021175384521s
Received healthy response to inference request in 1.2696306705474854s
5 requests
0 failed requests
5th percentile: 1.2701249599456788
10th percentile: 1.2706192493438722
20th percentile: 1.2716078281402587
30th percentile: 1.2723608016967773
40th percentile: 1.2728781700134277
50th percentile: 1.2733955383300781
60th percentile: 1.3136936187744142
70th percentile: 1.35399169921875
80th percentile: 2.093666362762452
90th percentile: 3.532717609405518
95th percentile: 4.25224323272705
99th percentile: 4.827863731384277
mean time: 2.0322075843811036
Pipeline stage StressChecker completed in 10.90s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.06s
grimjim-llama-3-nephilim_2191_v1 status is now deployed due to DeploymentManager action
grimjim-llama-3-nephilim_2191_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics