submission_id: chaiml-elo-alignment-run_5486_v1
developer_uid: zonemercy
status: inactive
model_repo: ChaiML/elo_alignment_run_3_ep1
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: 2024-07-12T13:32:49+00:00
model_name: dpo-v2-0710-rm-score-completed
model_group: ChaiML/elo_alignment_run
num_battles: 28117
num_wins: 16784
celo_rating: 1264.76
propriety_score: 0.7303042596348884
propriety_total_count: 12325.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: dpo-v2-0710-rm-score-completed
ineligible_reason: None
language_model: ChaiML/elo_alignment_run_3_ep1
model_size: 8B
reward_model: ChaiML/gpt2_xl_pairwise_89m_step_347634
us_pacific_date: 2024-07-12
win_ratio: 0.5969342390724472
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name chaiml-elo-alignment-run-5486-v1-mkmlizer
Waiting for job on chaiml-elo-alignment-run-5486-v1-mkmlizer to finish
chaiml-elo-alignment-run-5486-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ _____ __ __ ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ /___/ ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ Version: 0.8.14 ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ https://mk1.ai ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ The license key for the current software has been verified as ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ belonging to: ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ Chai Research Corp. ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ║ ║
chaiml-elo-alignment-run-5486-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-elo-alignment-run-5486-v1-mkmlizer: Downloaded to shared memory in 74.912s
chaiml-elo-alignment-run-5486-v1-mkmlizer: quantizing model to /dev/shm/model_cache
chaiml-elo-alignment-run-5486-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-elo-alignment-run-5486-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:06, 45.64it/s] Loading 0: 5%|▍ | 14/291 [00:00<00:04, 59.94it/s] Loading 0: 8%|▊ | 23/291 [00:00<00:04, 64.04it/s] Loading 0: 11%|█▏ | 33/291 [00:00<00:07, 36.57it/s] Loading 0: 13%|█▎ | 38/291 [00:00<00:06, 38.74it/s] Loading 0: 15%|█▍ | 43/291 [00:01<00:06, 40.40it/s] Loading 0: 17%|█▋ | 50/291 [00:01<00:05, 44.34it/s] Loading 0: 20%|██ | 59/291 [00:01<00:04, 51.38it/s] Loading 0: 23%|██▎ | 68/291 [00:01<00:03, 56.39it/s] Loading 0: 26%|██▌ | 76/291 [00:01<00:03, 62.04it/s] Loading 0: 29%|██▊ | 83/291 [00:01<00:06, 33.11it/s] Loading 0: 32%|███▏ | 92/291 [00:02<00:04, 40.14it/s] Loading 0: 35%|███▍ | 101/291 [00:02<00:04, 46.36it/s] Loading 0: 38%|███▊ | 110/291 [00:02<00:03, 51.17it/s] Loading 0: 42%|████▏ | 122/291 [00:02<00:02, 61.79it/s] Loading 0: 45%|████▍ | 130/291 [00:02<00:02, 63.04it/s] Loading 0: 47%|████▋ | 137/291 [00:03<00:04, 35.33it/s] Loading 0: 49%|████▉ | 143/291 [00:03<00:03, 38.00it/s] Loading 0: 51%|█████ | 149/291 [00:03<00:03, 40.49it/s] Loading 0: 54%|█████▍ | 157/291 [00:03<00:02, 47.81it/s] Loading 0: 57%|█████▋ | 166/291 [00:03<00:02, 53.14it/s] Loading 0: 60%|██████ | 175/291 [00:03<00:02, 57.25it/s] Loading 0: 64%|██████▍ | 186/291 [00:04<00:02, 39.20it/s] Loading 0: 66%|██████▌ | 192/291 [00:04<00:02, 39.88it/s] Loading 0: 69%|██████▉ | 201/291 [00:04<00:01, 45.72it/s] Loading 0: 72%|███████▏ | 210/291 [00:04<00:01, 50.60it/s] Loading 0: 75%|███████▌ | 219/291 [00:04<00:01, 54.99it/s] Loading 0: 78%|███████▊ | 227/291 [00:04<00:01, 52.76it/s] Loading 0: 80%|████████ | 233/291 [00:05<00:01, 33.35it/s] Loading 0: 82%|████████▏ | 238/291 [00:05<00:01, 35.65it/s] Loading 0: 85%|████████▍ | 247/291 [00:05<00:01, 42.96it/s] Loading 0: 88%|████████▊ | 256/291 [00:05<00:00, 49.24it/s] Loading 0: 91%|█████████ | 265/291 [00:05<00:00, 54.20it/s] Loading 0: 94%|█████████▍| 274/291 [00:05<00:00, 57.85it/s] Loading 0: 97%|█████████▋| 281/291 [00:05<00:00, 59.18it/s] Loading 0: 99%|█████████▉| 288/291 [00:12<00:00, 3.52it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
chaiml-elo-alignment-run-5486-v1-mkmlizer: quantized model in 31.661s
chaiml-elo-alignment-run-5486-v1-mkmlizer: Processed model ChaiML/elo_alignment_run_3_ep1 in 106.573s
chaiml-elo-alignment-run-5486-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-elo-alignment-run-5486-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-elo-alignment-run-5486-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-elo-alignment-run-5486-v1
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-elo-alignment-run-5486-v1/special_tokens_map.json
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-elo-alignment-run-5486-v1/config.json
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-elo-alignment-run-5486-v1/tokenizer_config.json
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-elo-alignment-run-5486-v1/tokenizer.json
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-elo-alignment-run-5486-v1/flywheel_model.0.safetensors
chaiml-elo-alignment-run-5486-v1-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
chaiml-elo-alignment-run-5486-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.
chaiml-elo-alignment-run-5486-v1-mkmlizer: warnings.warn(
chaiml-elo-alignment-run-5486-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`.
chaiml-elo-alignment-run-5486-v1-mkmlizer: warnings.warn(
chaiml-elo-alignment-run-5486-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.
chaiml-elo-alignment-run-5486-v1-mkmlizer: warnings.warn(
chaiml-elo-alignment-run-5486-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.
chaiml-elo-alignment-run-5486-v1-mkmlizer: warnings.warn(
chaiml-elo-alignment-run-5486-v1-mkmlizer: Downloading shards: 0%| | 0/2 [00:00<?, ?it/s] Downloading shards: 50%|█████ | 1/2 [00:07<00:07, 7.07s/it] Downloading shards: 100%|██████████| 2/2 [00:09<00:00, 4.35s/it] Downloading shards: 100%|██████████| 2/2 [00:09<00:00, 4.76s/it]
chaiml-elo-alignment-run-5486-v1-mkmlizer: Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.64it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 2.66it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 2.43it/s]
chaiml-elo-alignment-run-5486-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
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
Connection pool is full, discarding connection: %s
chaiml-elo-alignment-run-5486-v1-mkmlizer: Saving duration: 2.169s
chaiml-elo-alignment-run-5486-v1-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 15.020s
chaiml-elo-alignment-run-5486-v1-mkmlizer: creating bucket guanaco-reward-models
chaiml-elo-alignment-run-5486-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
chaiml-elo-alignment-run-5486-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/chaiml-elo-alignment-run-5486-v1_reward
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/chaiml-elo-alignment-run-5486-v1_reward/config.json
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/chaiml-elo-alignment-run-5486-v1_reward/tokenizer_config.json
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/chaiml-elo-alignment-run-5486-v1_reward/special_tokens_map.json
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/chaiml-elo-alignment-run-5486-v1_reward/merges.txt
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/chaiml-elo-alignment-run-5486-v1_reward/vocab.json
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/chaiml-elo-alignment-run-5486-v1_reward/tokenizer.json
chaiml-elo-alignment-run-5486-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/chaiml-elo-alignment-run-5486-v1_reward/reward.tensors
Job chaiml-elo-alignment-run-5486-v1-mkmlizer completed after 144.7s with status: succeeded
Stopping job with name chaiml-elo-alignment-run-5486-v1-mkmlizer
Pipeline stage MKMLizer completed in 145.70s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service chaiml-elo-alignment-run-5486-v1
Waiting for inference service chaiml-elo-alignment-run-5486-v1 to be ready
Inference service chaiml-elo-alignment-run-5486-v1 ready after 40.200974464416504s
Pipeline stage ISVCDeployer completed in 47.23s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.319734811782837s
Received healthy response to inference request in 1.570024013519287s
Received healthy response to inference request in 1.5691680908203125s
Received healthy response to inference request in 1.5173571109771729s
Received healthy response to inference request in 1.5939674377441406s
5 requests
0 failed requests
5th percentile: 1.5277193069458008
10th percentile: 1.5380815029144288
20th percentile: 1.5588058948516845
30th percentile: 1.5693392753601074
40th percentile: 1.5696816444396973
50th percentile: 1.570024013519287
60th percentile: 1.5796013832092286
70th percentile: 1.5891787528991699
80th percentile: 1.73912091255188
90th percentile: 2.0294278621673585
95th percentile: 2.1745813369750975
99th percentile: 2.290704116821289
mean time: 1.71405029296875
Pipeline stage StressChecker completed in 9.23s
chaiml-elo-alignment-run_5486_v1 status is now deployed due to DeploymentManager action
chaiml-elo-alignment-run_5486_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics