submission_id: chaiml-augment-alignment_v2
developer_uid: zonemercy
status: inactive
model_repo: ChaiML/augment-alignment
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: 2024-07-12T23:06:11+00:00
model_name: chaiml-augment-alignment_v2
model_group: ChaiML/augment-alignment
num_battles: 29055
num_wins: 17900
celo_rating: 1262.51
propriety_score: 0.7208114626188111
propriety_total_count: 7049.0
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-alignment_v2
ineligible_reason: None
language_model: ChaiML/augment-alignment
model_size: 8B
reward_model: ChaiML/gpt2_xl_pairwise_89m_step_347634
us_pacific_date: 2024-07-12
win_ratio: 0.6160729650662536
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name chaiml-augment-alignment-v2-mkmlizer
Waiting for job on chaiml-augment-alignment-v2-mkmlizer to finish
chaiml-augment-alignment-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-augment-alignment-v2-mkmlizer: ║ _____ __ __ ║
chaiml-augment-alignment-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
chaiml-augment-alignment-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
chaiml-augment-alignment-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
chaiml-augment-alignment-v2-mkmlizer: ║ /___/ ║
chaiml-augment-alignment-v2-mkmlizer: ║ ║
chaiml-augment-alignment-v2-mkmlizer: ║ Version: 0.8.14 ║
chaiml-augment-alignment-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
chaiml-augment-alignment-v2-mkmlizer: ║ https://mk1.ai ║
chaiml-augment-alignment-v2-mkmlizer: ║ ║
chaiml-augment-alignment-v2-mkmlizer: ║ The license key for the current software has been verified as ║
chaiml-augment-alignment-v2-mkmlizer: ║ belonging to: ║
chaiml-augment-alignment-v2-mkmlizer: ║ ║
chaiml-augment-alignment-v2-mkmlizer: ║ Chai Research Corp. ║
chaiml-augment-alignment-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaiml-augment-alignment-v2-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
chaiml-augment-alignment-v2-mkmlizer: ║ ║
chaiml-augment-alignment-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-augment-alignment-v2-mkmlizer: Downloaded to shared memory in 49.434s
chaiml-augment-alignment-v2-mkmlizer: quantizing model to /dev/shm/model_cache
chaiml-augment-alignment-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-augment-alignment-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:06, 43.83it/s] Loading 0: 5%|▍ | 14/291 [00:00<00:04, 58.10it/s] Loading 0: 8%|▊ | 22/291 [00:00<00:04, 66.09it/s] Loading 0: 11%|█▏ | 33/291 [00:00<00:07, 34.34it/s] Loading 0: 13%|█▎ | 38/291 [00:00<00:07, 36.10it/s] Loading 0: 15%|█▍ | 43/291 [00:01<00:06, 38.01it/s] Loading 0: 17%|█▋ | 50/291 [00:01<00:05, 42.03it/s] Loading 0: 20%|██ | 59/291 [00:01<00:04, 48.97it/s] Loading 0: 23%|██▎ | 68/291 [00:01<00:04, 54.08it/s] Loading 0: 26%|██▌ | 75/291 [00:01<00:03, 57.71it/s] Loading 0: 28%|██▊ | 82/291 [00:02<00:06, 32.45it/s] Loading 0: 31%|███▏ | 91/291 [00:02<00:05, 39.46it/s] Loading 0: 34%|███▍ | 100/291 [00:02<00:04, 45.65it/s] Loading 0: 37%|███▋ | 108/291 [00:02<00:03, 52.23it/s] Loading 0: 40%|███▉ | 115/291 [00:02<00:03, 55.65it/s] Loading 0: 42%|████▏ | 123/291 [00:02<00:02, 61.26it/s] Loading 0: 45%|████▍ | 130/291 [00:02<00:02, 59.56it/s] Loading 0: 47%|████▋ | 137/291 [00:03<00:04, 33.80it/s] Loading 0: 49%|████▉ | 143/291 [00:03<00:04, 36.85it/s] Loading 0: 51%|█████ | 149/291 [00:03<00:03, 39.64it/s] Loading 0: 54%|█████▍ | 157/291 [00:03<00:02, 47.47it/s] Loading 0: 57%|█████▋ | 165/291 [00:03<00:02, 52.94it/s] Loading 0: 60%|█████▉ | 174/291 [00:03<00:02, 56.78it/s] Loading 0: 62%|██████▏ | 181/291 [00:03<00:01, 59.67it/s] Loading 0: 65%|██████▍ | 188/291 [00:04<00:03, 32.91it/s] Loading 0: 66%|██████▋ | 193/291 [00:04<00:02, 35.07it/s] Loading 0: 69%|██████▉ | 201/291 [00:04<00:02, 43.08it/s] Loading 0: 72%|███████▏ | 210/291 [00:04<00:01, 49.32it/s] Loading 0: 75%|███████▌ | 219/291 [00:04<00:01, 54.23it/s] Loading 0: 78%|███████▊ | 227/291 [00:04<00:01, 55.61it/s] Loading 0: 80%|████████ | 234/291 [00:05<00:01, 33.28it/s] Loading 0: 82%|████████▏ | 239/291 [00:05<00:01, 35.15it/s] Loading 0: 85%|████████▌ | 248/291 [00:05<00:01, 42.31it/s] Loading 0: 88%|████████▊ | 256/291 [00:05<00:00, 49.44it/s] Loading 0: 91%|█████████ | 264/291 [00:05<00:00, 55.98it/s] Loading 0: 94%|█████████▍| 273/291 [00:05<00:00, 59.18it/s] Loading 0: 96%|█████████▌| 280/291 [00:06<00:00, 61.37it/s] Loading 0: 99%|█████████▊| 287/291 [00:13<00:01, 3.41it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
chaiml-augment-alignment-v2-mkmlizer: quantized model in 31.880s
chaiml-augment-alignment-v2-mkmlizer: Processed model ChaiML/augment-alignment in 81.314s
chaiml-augment-alignment-v2-mkmlizer: creating bucket guanaco-mkml-models
chaiml-augment-alignment-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-augment-alignment-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-augment-alignment-v2
chaiml-augment-alignment-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-augment-alignment-v2/config.json
chaiml-augment-alignment-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-augment-alignment-v2/special_tokens_map.json
chaiml-augment-alignment-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-augment-alignment-v2/tokenizer_config.json
chaiml-augment-alignment-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-augment-alignment-v2/tokenizer.json
chaiml-augment-alignment-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-augment-alignment-v2/flywheel_model.0.safetensors
chaiml-augment-alignment-v2-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
chaiml-augment-alignment-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-alignment-v2-mkmlizer: warnings.warn(
chaiml-augment-alignment-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-alignment-v2-mkmlizer: warnings.warn(
chaiml-augment-alignment-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-alignment-v2-mkmlizer: warnings.warn(
chaiml-augment-alignment-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-alignment-v2-mkmlizer: warnings.warn(
chaiml-augment-alignment-v2-mkmlizer: Downloading shards: 0%| | 0/2 [00:00<?, ?it/s] Downloading shards: 50%|█████ | 1/2 [00:06<00:06, 6.81s/it] Downloading shards: 100%|██████████| 2/2 [00:09<00:00, 4.17s/it] Downloading shards: 100%|██████████| 2/2 [00:09<00:00, 4.56s/it]
chaiml-augment-alignment-v2-mkmlizer: Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.61it/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-alignment-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
chaiml-augment-alignment-v2-mkmlizer: Saving duration: 2.136s
chaiml-augment-alignment-v2-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 13.770s
chaiml-augment-alignment-v2-mkmlizer: creating bucket guanaco-reward-models
chaiml-augment-alignment-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
chaiml-augment-alignment-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/chaiml-augment-alignment-v2_reward
chaiml-augment-alignment-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/chaiml-augment-alignment-v2_reward/config.json
chaiml-augment-alignment-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/chaiml-augment-alignment-v2_reward/merges.txt
chaiml-augment-alignment-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/chaiml-augment-alignment-v2_reward/tokenizer_config.json
chaiml-augment-alignment-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/chaiml-augment-alignment-v2_reward/special_tokens_map.json
chaiml-augment-alignment-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/chaiml-augment-alignment-v2_reward/vocab.json
chaiml-augment-alignment-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/chaiml-augment-alignment-v2_reward/tokenizer.json
chaiml-augment-alignment-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/chaiml-augment-alignment-v2_reward/reward.tensors
Job chaiml-augment-alignment-v2-mkmlizer completed after 128.1s with status: succeeded
Stopping job with name chaiml-augment-alignment-v2-mkmlizer
Pipeline stage MKMLizer completed in 128.97s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service chaiml-augment-alignment-v2
Waiting for inference service chaiml-augment-alignment-v2 to be ready
Inference service chaiml-augment-alignment-v2 ready after 50.31722831726074s
Pipeline stage ISVCDeployer completed in 59.36s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.622286796569824s
Received healthy response to inference request in 4.178320646286011s
Received healthy response to inference request in 1.543539047241211s
Received healthy response to inference request in 1.5098979473114014s
Received healthy response to inference request in 1.8401587009429932s
5 requests
0 failed requests
5th percentile: 1.5166261672973633
10th percentile: 1.5233543872833253
20th percentile: 1.536810827255249
30th percentile: 1.6028629779815673
40th percentile: 1.7215108394622802
50th percentile: 1.8401587009429932
60th percentile: 2.1530099391937254
70th percentile: 2.4658611774444577
80th percentile: 2.9334935665130617
90th percentile: 3.555907106399536
95th percentile: 3.8671138763427733
99th percentile: 4.116079292297363
mean time: 2.338840627670288
Pipeline stage StressChecker completed in 12.41s
chaiml-augment-alignment_v2 status is now deployed due to DeploymentManager action
chaiml-augment-alignment_v2 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics