submission_id: chaiml-augment-ensemble_v2
developer_uid: zonemercy
status: inactive
model_repo: ChaiML/augment-ensemble
reward_repo: ChaiML/gpt2_xl_pairwise_89m_step_347634
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.08, 'top_k': 40, '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: 2024-07-12T14:07:23+00:00
model_name: chaiml-augment-ensemble_v1
model_group: ChaiML/augment-ensemble
num_battles: 17459
num_wins: 10408
celo_rating: 1269.01
alignment_score: None
alignment_samples: 0
propriety_score: 0.7140786980269758
propriety_total_count: 8971.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: chaiml-augment-ensemble_v1
ineligible_reason: None
language_model: ChaiML/augment-ensemble
model_size: 8B
reward_model: ChaiML/gpt2_xl_pairwise_89m_step_347634
us_pacific_date: 2024-07-12
win_ratio: 0.5961395268915746
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name chaiml-augment-ensemble-v2-mkmlizer
Waiting for job on chaiml-augment-ensemble-v2-mkmlizer to finish
chaiml-augment-ensemble-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-augment-ensemble-v2-mkmlizer: ║ _____ __ __ ║
chaiml-augment-ensemble-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
chaiml-augment-ensemble-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
chaiml-augment-ensemble-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
chaiml-augment-ensemble-v2-mkmlizer: ║ /___/ ║
chaiml-augment-ensemble-v2-mkmlizer: ║ ║
chaiml-augment-ensemble-v2-mkmlizer: ║ Version: 0.8.14 ║
chaiml-augment-ensemble-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
chaiml-augment-ensemble-v2-mkmlizer: ║ https://mk1.ai ║
chaiml-augment-ensemble-v2-mkmlizer: ║ ║
chaiml-augment-ensemble-v2-mkmlizer: ║ The license key for the current software has been verified as ║
chaiml-augment-ensemble-v2-mkmlizer: ║ belonging to: ║
chaiml-augment-ensemble-v2-mkmlizer: ║ ║
chaiml-augment-ensemble-v2-mkmlizer: ║ Chai Research Corp. ║
chaiml-augment-ensemble-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaiml-augment-ensemble-v2-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
chaiml-augment-ensemble-v2-mkmlizer: ║ ║
chaiml-augment-ensemble-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
chaiml-augment-ensemble-v2-mkmlizer: Downloaded to shared memory in 50.038s
chaiml-augment-ensemble-v2-mkmlizer: quantizing model to /dev/shm/model_cache
chaiml-augment-ensemble-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-augment-ensemble-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:06, 44.31it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:04, 64.05it/s] Loading 0: 8%|▊ | 22/291 [00:00<00:04, 65.26it/s] Loading 0: 11%|█▏ | 33/291 [00:00<00:07, 34.62it/s] Loading 0: 13%|█▎ | 38/291 [00:00<00:06, 36.83it/s] Loading 0: 15%|█▍ | 43/291 [00:01<00:06, 38.65it/s] Loading 0: 17%|█▋ | 50/291 [00:01<00:05, 42.73it/s] Loading 0: 20%|██ | 59/291 [00:01<00:04, 49.73it/s] Loading 0: 23%|██▎ | 68/291 [00:01<00:04, 54.95it/s] Loading 0: 26%|██▌ | 76/291 [00:01<00:03, 60.67it/s] Loading 0: 29%|██▊ | 83/291 [00:02<00:06, 32.33it/s] Loading 0: 31%|███▏ | 91/291 [00:02<00:05, 39.72it/s] Loading 0: 34%|███▍ | 100/291 [00:02<00:04, 46.22it/s] Loading 0: 37%|███▋ | 109/291 [00:02<00:03, 51.39it/s] Loading 0: 40%|███▉ | 116/291 [00:02<00:03, 55.18it/s] Loading 0: 43%|████▎ | 126/291 [00:02<00:02, 60.89it/s] Loading 0: 46%|████▌ | 133/291 [00:03<00:04, 36.27it/s] Loading 0: 48%|████▊ | 139/291 [00:03<00:04, 36.41it/s] Loading 0: 51%|█████ | 148/291 [00:03<00:03, 43.28it/s] Loading 0: 54%|█████▍ | 157/291 [00:03<00:02, 48.70it/s] Loading 0: 57%|█████▋ | 166/291 [00:03<00:02, 53.31it/s] Loading 0: 60%|██████ | 175/291 [00:03<00:02, 56.92it/s] Loading 0: 64%|██████▍ | 186/291 [00:04<00:02, 39.14it/s] Loading 0: 66%|██████▌ | 192/291 [00:04<00:02, 39.81it/s] Loading 0: 69%|██████▉ | 201/291 [00:04<00:01, 45.83it/s] Loading 0: 72%|███████▏ | 210/291 [00:04<00:01, 50.92it/s] Loading 0: 75%|███████▌ | 219/291 [00:04<00:01, 55.36it/s] Loading 0: 78%|███████▊ | 227/291 [00:04<00:01, 56.79it/s] Loading 0: 80%|████████ | 234/291 [00:05<00:01, 34.36it/s] Loading 0: 82%|████████▏ | 239/291 [00:05<00:01, 36.47it/s] Loading 0: 85%|████████▌ | 248/291 [00:05<00:00, 43.53it/s] Loading 0: 88%|████████▊ | 256/291 [00:05<00:00, 50.47it/s] Loading 0: 91%|█████████ | 265/291 [00:05<00:00, 55.05it/s] Loading 0: 94%|█████████▍| 274/291 [00:05<00:00, 58.77it/s] Loading 0: 97%|█████████▋| 283/291 [00:05<00:00, 64.52it/s] Loading 0: 100%|█████████▉| 290/291 [00:12<00:00, 3.70it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
chaiml-augment-ensemble-v2-mkmlizer: quantized model in 32.753s
chaiml-augment-ensemble-v2-mkmlizer: Processed model ChaiML/augment-ensemble in 82.791s
chaiml-augment-ensemble-v2-mkmlizer: creating bucket guanaco-mkml-models
chaiml-augment-ensemble-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-augment-ensemble-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-augment-ensemble-v2
chaiml-augment-ensemble-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-augment-ensemble-v2/special_tokens_map.json
chaiml-augment-ensemble-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-augment-ensemble-v2/tokenizer_config.json
chaiml-augment-ensemble-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-augment-ensemble-v2/config.json
chaiml-augment-ensemble-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-augment-ensemble-v2/tokenizer.json
chaiml-augment-ensemble-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-augment-ensemble-v2/flywheel_model.0.safetensors
chaiml-augment-ensemble-v2-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
chaiml-augment-ensemble-v2-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.
chaiml-augment-ensemble-v2-mkmlizer: warnings.warn(
chaiml-augment-ensemble-v2-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`.
chaiml-augment-ensemble-v2-mkmlizer: warnings.warn(
chaiml-augment-ensemble-v2-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.
chaiml-augment-ensemble-v2-mkmlizer: warnings.warn(
chaiml-augment-ensemble-v2-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.
chaiml-augment-ensemble-v2-mkmlizer: warnings.warn(
chaiml-augment-ensemble-v2-mkmlizer: Downloading shards: 0%| | 0/2 [00:00<?, ?it/s] Downloading shards: 50%|█████ | 1/2 [00:06<00:06, 6.77s/it] Downloading shards: 100%|██████████| 2/2 [00:09<00:00, 4.27s/it] Downloading shards: 100%|██████████| 2/2 [00:09<00:00, 4.65s/it]
chaiml-augment-ensemble-v2-mkmlizer: Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.59it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 2.65it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 2.41it/s]
chaiml-augment-ensemble-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
chaiml-augment-ensemble-v2-mkmlizer: Saving duration: 2.207s
chaiml-augment-ensemble-v2-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 14.921s
chaiml-augment-ensemble-v2-mkmlizer: creating bucket guanaco-reward-models
chaiml-augment-ensemble-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
chaiml-augment-ensemble-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/chaiml-augment-ensemble-v2_reward
chaiml-augment-ensemble-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/chaiml-augment-ensemble-v2_reward/special_tokens_map.json
chaiml-augment-ensemble-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/chaiml-augment-ensemble-v2_reward/config.json
chaiml-augment-ensemble-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/chaiml-augment-ensemble-v2_reward/vocab.json
chaiml-augment-ensemble-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/chaiml-augment-ensemble-v2_reward/tokenizer_config.json
chaiml-augment-ensemble-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/chaiml-augment-ensemble-v2_reward/tokenizer.json
chaiml-augment-ensemble-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/chaiml-augment-ensemble-v2_reward/merges.txt
chaiml-augment-ensemble-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/chaiml-augment-ensemble-v2_reward/reward.tensors
Job chaiml-augment-ensemble-v2-mkmlizer completed after 124.98s with status: succeeded
Stopping job with name chaiml-augment-ensemble-v2-mkmlizer
Pipeline stage MKMLizer completed in 125.99s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service chaiml-augment-ensemble-v2
Waiting for inference service chaiml-augment-ensemble-v2 to be ready
Inference service chaiml-augment-ensemble-v2 ready after 40.33527445793152s
Pipeline stage ISVCDeployer completed in 47.06s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.28757381439209s
Received healthy response to inference request in 1.5618374347686768s
Received healthy response to inference request in 1.538069486618042s
Received healthy response to inference request in 1.5113511085510254s
Received healthy response to inference request in 1.586212396621704s
5 requests
0 failed requests
5th percentile: 1.5166947841644287
10th percentile: 1.522038459777832
20th percentile: 1.5327258110046387
30th percentile: 1.542823076248169
40th percentile: 1.5523302555084229
50th percentile: 1.5618374347686768
60th percentile: 1.5715874195098878
70th percentile: 1.5813374042510986
80th percentile: 1.7264846801757814
90th percentile: 2.0070292472839357
95th percentile: 2.1473015308380123
99th percentile: 2.2595193576812744
mean time: 1.6970088481903076
Pipeline stage StressChecker completed in 9.11s
chaiml-augment-ensemble_v2 status is now deployed due to DeploymentManager action
chaiml-augment-ensemble_v2 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics