submission_id: chaiml-elo-alignment-run-3_v22
developer_uid: robert_irvine
alignment_samples: 0
best_of: 4
celo_rating: 1228.62
display_name: chaiml-elo-alignment-run-3_v22
formatter: {'memory_template': '\n####\n', 'prompt_template': '<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
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': 4, 'max_output_tokens': 64, 'reward_max_token_input': 256}
is_internal_developer: True
language_model: ChaiML/elo_alignment_run_3
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: ChaiML/elo_alignment_run
model_name: chaiml-elo-alignment-run-3_v22
model_num_parameters: 8030261248.0
model_repo: ChaiML/elo_alignment_run_3
model_size: 8B
num_battles: 164172
num_wins: 89072
propriety_score: 0.7247523390203633
propriety_total_count: 14536.0
ranking_group: single
reward_formatter: {'bot_template': 'Bot: {message}\n', 'memory_template': '', 'prompt_template': '', 'response_template': 'Bot:', 'truncate_by_message': False, 'user_template': 'User: {message}\n'}
reward_repo: ChaiML/gpt2_xl_pairwise_89m_step_347634
status: torndown
submission_type: basic
timestamp: 2024-08-05T01:23:12+00:00
us_pacific_date: 2024-08-04
win_ratio: 0.5425529322905246
Download Preference Data
Resubmit model
Running pipeline stage MKMLizer
Starting job with name chaiml-elo-alignment-run-3-v22-mkmlizer
Waiting for job on chaiml-elo-alignment-run-3-v22-mkmlizer to finish
chaiml-elo-alignment-run-3-v22-mkmlizer: Downloaded to shared memory in 41.388s
chaiml-elo-alignment-run-3-v22-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp3v7thigt, device:0
chaiml-elo-alignment-run-3-v22-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Connection pool is full, discarding connection: %s. Connection pool size: %s
chaiml-elo-alignment-run-3-v22-mkmlizer: quantized model in 29.276s
chaiml-elo-alignment-run-3-v22-mkmlizer: Processed model ChaiML/elo_alignment_run_3 in 70.665s
chaiml-elo-alignment-run-3-v22-mkmlizer: creating bucket guanaco-mkml-models
chaiml-elo-alignment-run-3-v22-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-elo-alignment-run-3-v22-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-elo-alignment-run-3-v22
chaiml-elo-alignment-run-3-v22-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-elo-alignment-run-3-v22/tokenizer_config.json
chaiml-elo-alignment-run-3-v22-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-elo-alignment-run-3-v22/tokenizer.json
chaiml-elo-alignment-run-3-v22-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-elo-alignment-run-3-v22/flywheel_model.0.safetensors
chaiml-elo-alignment-run-3-v22-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
chaiml-elo-alignment-run-3-v22-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:11, 24.54it/s] Loading 0: 4%|▍ | 12/291 [00:00<00:08, 33.43it/s] Loading 0: 5%|▌ | 16/291 [00:00<00:08, 30.63it/s] Loading 0: 7%|▋ | 21/291 [00:00<00:08, 32.86it/s] Loading 0: 9%|▊ | 25/291 [00:00<00:08, 30.68it/s] Loading 0: 10%|█ | 30/291 [00:00<00:07, 34.60it/s] Loading 0: 12%|█▏ | 34/291 [00:01<00:12, 21.40it/s] Loading 0: 13%|█▎ | 37/291 [00:01<00:11, 21.44it/s] Loading 0: 14%|█▍ | 41/291 [00:01<00:11, 21.60it/s] Loading 0: 16%|█▋ | 48/291 [00:01<00:08, 28.83it/s] Loading 0: 18%|█▊ | 52/291 [00:01<00:08, 28.75it/s] Loading 0: 20%|█▉ | 57/291 [00:02<00:07, 31.33it/s] Loading 0: 21%|██ | 61/291 [00:02<00:07, 30.77it/s] Loading 0: 23%|██▎ | 66/291 [00:02<00:06, 33.15it/s] Loading 0: 24%|██▍ | 70/291 [00:02<00:07, 31.48it/s] Loading 0: 25%|██▌ | 74/291 [00:02<00:07, 30.48it/s] Loading 0: 27%|██▋ | 78/291 [00:02<00:06, 30.74it/s] Loading 0: 28%|██▊ | 82/291 [00:03<00:09, 21.00it/s] Loading 0: 29%|██▉ | 85/291 [00:03<00:09, 22.30it/s] Loading 0: 31%|███ | 90/291 [00:03<00:07, 26.39it/s] Loading 0: 32%|███▏ | 94/291 [00:03<00:07, 26.54it/s] Loading 0: 34%|███▍ | 99/291 [00:03<00:06, 29.99it/s] Loading 0: 35%|███▌ | 103/291 [00:03<00:06, 29.66it/s] Loading 0: 37%|███▋ | 108/291 [00:03<00:05, 32.18it/s] Loading 0: 38%|███▊ | 112/291 [00:03<00:05, 31.59it/s] Loading 0: 40%|███▉ | 116/291 [00:04<00:05, 32.45it/s] Loading 0: 42%|████▏ | 122/291 [00:04<00:04, 37.05it/s] Loading 0: 44%|████▎ | 127/291 [00:04<00:04, 34.42it/s] Loading 0: 46%|████▌ | 133/291 [00:04<00:05, 28.59it/s] Loading 0: 47%|████▋ | 137/291 [00:04<00:05, 28.76it/s] Loading 0: 48%|████▊ | 141/291 [00:04<00:05, 27.25it/s] Loading 0: 51%|█████ | 147/291 [00:05<00:04, 31.71it/s] Loading 0: 52%|█████▏ | 151/291 [00:05<00:04, 30.85it/s] Loading 0: 54%|█████▎ | 156/291 [00:05<00:04, 32.59it/s] Loading 0: 55%|█████▍ | 160/291 [00:05<00:04, 30.31it/s] Loading 0: 57%|█████▋ | 165/291 [00:05<00:03, 32.58it/s] Loading 0: 58%|█████▊ | 169/291 [00:05<00:03, 31.60it/s] Loading 0: 60%|█████▉ | 174/291 [00:05<00:03, 33.76it/s] Loading 0: 61%|██████ | 178/291 [00:06<00:03, 32.81it/s] Loading 0: 63%|██████▎ | 184/291 [00:06<00:02, 38.94it/s] Loading 0: 65%|██████▍ | 189/291 [00:06<00:04, 24.56it/s] Loading 0: 67%|██████▋ | 194/291 [00:06<00:03, 26.46it/s] Loading 0: 69%|██████▉ | 201/291 [00:06<00:02, 33.02it/s] Loading 0: 70%|███████ | 205/291 [00:06<00:02, 32.53it/s] Loading 0: 72%|███████▏ | 210/291 [00:07<00:02, 35.54it/s] Loading 0: 74%|███████▎ | 214/291 [00:07<00:02, 34.54it/s] Loading 0: 75%|███████▌ | 219/291 [00:07<00:01, 36.71it/s] Loading 0: 77%|███████▋ | 223/291 [00:07<00:01, 35.29it/s] Loading 0: 78%|███████▊ | 227/291 [00:07<00:01, 35.19it/s] Loading 0: 79%|███████▉ | 231/291 [00:07<00:01, 35.20it/s] Loading 0: 81%|████████ | 235/291 [00:07<00:02, 25.76it/s] Loading 0: 82%|████████▏ | 239/291 [00:08<00:02, 24.83it/s] Loading 0: 85%|████████▍ | 246/291 [00:08<00:01, 32.32it/s] Loading 0: 86%|████████▌ | 250/291 [00:08<00:01, 32.27it/s] Loading 0: 88%|████████▊ | 255/291 [00:08<00:01, 35.45it/s] Loading 0: 89%|████████▉ | 259/291 [00:08<00:00, 34.49it/s] Loading 0: 91%|█████████ | 264/291 [00:08<00:00, 35.44it/s] Loading 0: 92%|█████████▏| 268/291 [00:08<00:00, 33.85it/s] Loading 0: 94%|█████████▍| 273/291 [00:08<00:00, 36.27it/s] Loading 0: 95%|█████████▌| 277/291 [00:09<00:00, 34.34it/s] Loading 0: 97%|█████████▋| 281/291 [00:09<00:00, 34.58it/s] Loading 0: 98%|█████████▊| 286/291 [00:14<00:01, 2.60it/s] Loading 0: 99%|█████████▉| 289/291 [00:14<00:00, 3.23it/s] /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:957: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
chaiml-elo-alignment-run-3-v22-mkmlizer: warnings.warn(
chaiml-elo-alignment-run-3-v22-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:785: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
chaiml-elo-alignment-run-3-v22-mkmlizer: warnings.warn(
chaiml-elo-alignment-run-3-v22-mkmlizer: Downloading shards: 0%| | 0/2 [00:00<?, ?it/s] Downloading shards: 50%|█████ | 1/2 [00:05<00:05, 5.54s/it] Downloading shards: 100%|██████████| 2/2 [00:08<00:00, 3.87s/it] Downloading shards: 100%|██████████| 2/2 [00:08<00:00, 4.12s/it]
chaiml-elo-alignment-run-3-v22-mkmlizer: Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 2.45it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 4.06it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 3.70it/s]
chaiml-elo-alignment-run-3-v22-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
chaiml-elo-alignment-run-3-v22-mkmlizer: Saving duration: 1.345s
chaiml-elo-alignment-run-3-v22-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 13.119s
chaiml-elo-alignment-run-3-v22-mkmlizer: creating bucket guanaco-reward-models
chaiml-elo-alignment-run-3-v22-mkmlizer: Bucket 's3://guanaco-reward-models/' created
chaiml-elo-alignment-run-3-v22-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v22_reward
chaiml-elo-alignment-run-3-v22-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v22_reward/config.json
chaiml-elo-alignment-run-3-v22-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v22_reward/special_tokens_map.json
chaiml-elo-alignment-run-3-v22-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v22_reward/tokenizer_config.json
chaiml-elo-alignment-run-3-v22-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v22_reward/merges.txt
chaiml-elo-alignment-run-3-v22-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v22_reward/vocab.json
chaiml-elo-alignment-run-3-v22-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v22_reward/tokenizer.json
chaiml-elo-alignment-run-3-v22-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v22_reward/reward.tensors
Job chaiml-elo-alignment-run-3-v22-mkmlizer completed after 126.43s with status: succeeded
Stopping job with name chaiml-elo-alignment-run-3-v22-mkmlizer
Pipeline stage MKMLizer completed in 127.49s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.08s
Running pipeline stage ISVCDeployer
Creating inference service chaiml-elo-alignment-run-3-v22
Waiting for inference service chaiml-elo-alignment-run-3-v22 to be ready
Inference service chaiml-elo-alignment-run-3-v22 ready after 161.00603413581848s
Pipeline stage ISVCDeployer completed in 162.80s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.8371732234954834s
Received healthy response to inference request in 0.9022696018218994s
Received healthy response to inference request in 0.9891932010650635s
Received healthy response to inference request in 0.9787828922271729s
Received healthy response to inference request in 1.0012543201446533s
5 requests
0 failed requests
5th percentile: 0.9175722599029541
10th percentile: 0.9328749179840088
20th percentile: 0.9634802341461182
30th percentile: 0.9808649539947509
40th percentile: 0.9850290775299072
50th percentile: 0.9891932010650635
60th percentile: 0.9940176486968995
70th percentile: 0.9988420963287353
80th percentile: 1.1684381008148195
90th percentile: 1.5028056621551515
95th percentile: 1.6699894428253172
99th percentile: 1.8037364673614502
mean time: 1.1417346477508545
Pipeline stage StressChecker completed in 6.32s
chaiml-elo-alignment-run-3_v22 status is now deployed due to DeploymentManager action
chaiml-elo-alignment-run-3_v22 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of chaiml-elo-alignment-run-3_v22
Running pipeline stage ISVCDeleter
Checking if service chaiml-elo-alignment-run-3-v22 is running
Tearing down inference service chaiml-elo-alignment-run-3-v22
Service chaiml-elo-alignment-run-3-v22 has been torndown
Pipeline stage ISVCDeleter completed in 5.54s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key chaiml-elo-alignment-run-3-v22/config.json from bucket guanaco-mkml-models
Deleting key chaiml-elo-alignment-run-3-v22/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key chaiml-elo-alignment-run-3-v22/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key chaiml-elo-alignment-run-3-v22/tokenizer.json from bucket guanaco-mkml-models
Deleting key chaiml-elo-alignment-run-3-v22/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key chaiml-elo-alignment-run-3-v22_reward/config.json from bucket guanaco-reward-models
Deleting key chaiml-elo-alignment-run-3-v22_reward/merges.txt from bucket guanaco-reward-models
Deleting key chaiml-elo-alignment-run-3-v22_reward/reward.tensors from bucket guanaco-reward-models
Deleting key chaiml-elo-alignment-run-3-v22_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key chaiml-elo-alignment-run-3-v22_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key chaiml-elo-alignment-run-3-v22_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key chaiml-elo-alignment-run-3-v22_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.53s
chaiml-elo-alignment-run-3_v22 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics