submission_id: jic062-share-qlora-out_v7
developer_uid: chace9580
alignment_samples: 0
best_of: 16
celo_rating: 1163.92
display_name: jic062-share-qlora-out_v7
formatter: {'memory_template': '<|im_start|>system\n{memory}<|im_end|>\n', 'prompt_template': '<|im_start|>user\n{prompt}<|im_end|>\n', 'bot_template': '<|im_start|>assistant\n{bot_name}: {message}<|im_end|>\n', 'user_template': '<|im_start|>user\n{user_name}: {message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n{bot_name}:', 'truncate_by_message': False}
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64, 'reward_max_token_input': 1024}
is_internal_developer: False
language_model: jic062/share-qlora-out
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: jic062/share-qlora-out
model_name: jic062-share-qlora-out_v7
model_num_parameters: 8030261248.0
model_repo: jic062/share-qlora-out
model_size: 8B
num_battles: 11570
num_wins: 5156
propriety_score: 0.6859344894026975
propriety_total_count: 1038.0
ranking_group: single
reward_formatter: {'bot_template': '<|im_start|>assistant\n{bot_name}: {message}<|im_end|>\n', 'memory_template': '<|im_start|>system\n{memory}<|im_end|>\n', 'prompt_template': '<|im_start|>user\n{prompt}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n{bot_name}:', 'truncate_by_message': False, 'user_template': '<|im_start|>user\n{user_name}: {message}<|im_end|>\n'}
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
status: torndown
submission_type: basic
timestamp: 2024-07-26T07:43:25+00:00
us_pacific_date: 2024-07-26
win_ratio: 0.4456352636127917
Download Preference Data
Resubmit model
Running pipeline stage MKMLizer
Starting job with name jic062-share-qlora-out-v7-mkmlizer
Waiting for job on jic062-share-qlora-out-v7-mkmlizer to finish
jic062-share-qlora-out-v7-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-share-qlora-out-v7-mkmlizer: ║ _____ __ __ ║
jic062-share-qlora-out-v7-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
jic062-share-qlora-out-v7-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
jic062-share-qlora-out-v7-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
jic062-share-qlora-out-v7-mkmlizer: ║ /___/ ║
jic062-share-qlora-out-v7-mkmlizer: ║ ║
jic062-share-qlora-out-v7-mkmlizer: ║ Version: 0.9.7 ║
jic062-share-qlora-out-v7-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
jic062-share-qlora-out-v7-mkmlizer: ║ https://mk1.ai ║
jic062-share-qlora-out-v7-mkmlizer: ║ ║
jic062-share-qlora-out-v7-mkmlizer: ║ The license key for the current software has been verified as ║
jic062-share-qlora-out-v7-mkmlizer: ║ belonging to: ║
jic062-share-qlora-out-v7-mkmlizer: ║ ║
jic062-share-qlora-out-v7-mkmlizer: ║ Chai Research Corp. ║
jic062-share-qlora-out-v7-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
jic062-share-qlora-out-v7-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
jic062-share-qlora-out-v7-mkmlizer: ║ ║
jic062-share-qlora-out-v7-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-share-qlora-out-v7-mkmlizer: Downloaded to shared memory in 25.654s
jic062-share-qlora-out-v7-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmptpx21xj_, device:0
jic062-share-qlora-out-v7-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-share-qlora-out-v7-mkmlizer: quantized model in 26.129s
jic062-share-qlora-out-v7-mkmlizer: Processed model jic062/share-qlora-out in 51.783s
jic062-share-qlora-out-v7-mkmlizer: creating bucket guanaco-mkml-models
jic062-share-qlora-out-v7-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-share-qlora-out-v7-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-share-qlora-out-v7
jic062-share-qlora-out-v7-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-share-qlora-out-v7/config.json
jic062-share-qlora-out-v7-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-share-qlora-out-v7/special_tokens_map.json
jic062-share-qlora-out-v7-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-share-qlora-out-v7/tokenizer_config.json
jic062-share-qlora-out-v7-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-share-qlora-out-v7/tokenizer.json
jic062-share-qlora-out-v7-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-share-qlora-out-v7/flywheel_model.0.safetensors
jic062-share-qlora-out-v7-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
jic062-share-qlora-out-v7-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 7/291 [00:00<00:05, 48.59it/s] Loading 0: 5%|▌ | 16/291 [00:00<00:04, 65.16it/s] Loading 0: 9%|▊ | 25/291 [00:00<00:03, 72.35it/s] Loading 0: 12%|█▏ | 34/291 [00:00<00:03, 73.87it/s] Loading 0: 15%|█▍ | 43/291 [00:00<00:03, 77.25it/s] Loading 0: 18%|█▊ | 52/291 [00:00<00:03, 79.49it/s] Loading 0: 21%|██ | 61/291 [00:00<00:02, 80.93it/s] Loading 0: 24%|██▍ | 70/291 [00:00<00:02, 77.88it/s] Loading 0: 27%|██▋ | 79/291 [00:01<00:02, 79.44it/s] Loading 0: 30%|██▉ | 87/291 [00:02<00:10, 20.37it/s] Loading 0: 32%|███▏ | 94/291 [00:02<00:07, 24.91it/s] Loading 0: 35%|███▌ | 103/291 [00:02<00:05, 32.54it/s] Loading 0: 38%|███▊ | 112/291 [00:02<00:04, 40.40it/s] Loading 0: 42%|████▏ | 121/291 [00:02<00:03, 46.82it/s] Loading 0: 45%|████▍ | 130/291 [00:02<00:02, 53.68it/s] Loading 0: 48%|████▊ | 139/291 [00:02<00:02, 59.02it/s] Loading 0: 51%|█████ | 148/291 [00:02<00:02, 65.41it/s] Loading 0: 54%|█████▍ | 157/291 [00:03<00:01, 70.20it/s] Loading 0: 57%|█████▋ | 166/291 [00:03<00:01, 70.55it/s] Loading 0: 60%|██████ | 175/291 [00:03<00:01, 75.27it/s] Loading 0: 63%|██████▎ | 184/291 [00:03<00:01, 79.11it/s] Loading 0: 66%|██████▋ | 193/291 [00:04<00:04, 21.31it/s] Loading 0: 69%|██████▉ | 202/291 [00:04<00:03, 27.61it/s] Loading 0: 73%|███████▎ | 211/291 [00:04<00:02, 34.71it/s] Loading 0: 76%|███████▌ | 220/291 [00:04<00:01, 42.18it/s] Loading 0: 79%|███████▊ | 229/291 [00:04<00:01, 49.35it/s] Loading 0: 82%|████████▏ | 238/291 [00:05<00:00, 55.70it/s] Loading 0: 85%|████████▍ | 247/291 [00:05<00:00, 61.86it/s] Loading 0: 88%|████████▊ | 256/291 [00:05<00:00, 67.13it/s] Loading 0: 91%|█████████ | 265/291 [00:05<00:00, 66.76it/s] Loading 0: 94%|█████████▍| 274/291 [00:05<00:00, 71.78it/s] Loading 0: 98%|█████████▊| 284/291 [00:05<00:00, 78.84it/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.
jic062-share-qlora-out-v7-mkmlizer: warnings.warn(
jic062-share-qlora-out-v7-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.
jic062-share-qlora-out-v7-mkmlizer: warnings.warn(
jic062-share-qlora-out-v7-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:469: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jic062-share-qlora-out-v7-mkmlizer: warnings.warn(
jic062-share-qlora-out-v7-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jic062-share-qlora-out-v7-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jic062-share-qlora-out-v7-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jic062-share-qlora-out-v7_reward
jic062-share-qlora-out-v7-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jic062-share-qlora-out-v7_reward/config.json
jic062-share-qlora-out-v7-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jic062-share-qlora-out-v7_reward/special_tokens_map.json
jic062-share-qlora-out-v7-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jic062-share-qlora-out-v7_reward/tokenizer_config.json
jic062-share-qlora-out-v7-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jic062-share-qlora-out-v7_reward/merges.txt
jic062-share-qlora-out-v7-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jic062-share-qlora-out-v7_reward/vocab.json
jic062-share-qlora-out-v7-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jic062-share-qlora-out-v7_reward/tokenizer.json
jic062-share-qlora-out-v7-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jic062-share-qlora-out-v7_reward/reward.tensors
Job jic062-share-qlora-out-v7-mkmlizer completed after 86.43s with status: succeeded
Stopping job with name jic062-share-qlora-out-v7-mkmlizer
Pipeline stage MKMLizer completed in 87.42s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service jic062-share-qlora-out-v7
Waiting for inference service jic062-share-qlora-out-v7 to be ready
Inference service jic062-share-qlora-out-v7 ready after 80.62507271766663s
Pipeline stage ISVCDeployer completed in 83.03s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.293663740158081s
Received healthy response to inference request in 1.373905897140503s
Received healthy response to inference request in 1.3666431903839111s
Received healthy response to inference request in 1.459681749343872s
Received healthy response to inference request in 1.369767665863037s
5 requests
0 failed requests
5th percentile: 1.3672680854797363
10th percentile: 1.3678929805755615
20th percentile: 1.369142770767212
30th percentile: 1.3705953121185304
40th percentile: 1.3722506046295166
50th percentile: 1.373905897140503
60th percentile: 1.4082162380218506
70th percentile: 1.4425265789031982
80th percentile: 1.626478147506714
90th percentile: 1.9600709438323975
95th percentile: 2.126867341995239
99th percentile: 2.2603044605255125
mean time: 1.5727324485778809
Pipeline stage StressChecker completed in 8.55s
jic062-share-qlora-out_v7 status is now deployed due to DeploymentManager action
jic062-share-qlora-out_v7 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of jic062-share-qlora-out_v7
Running pipeline stage ISVCDeleter
Checking if service jic062-share-qlora-out-v7 is running
Tearing down inference service jic062-share-qlora-out-v7
Service jic062-share-qlora-out-v7 has been torndown
Pipeline stage ISVCDeleter completed in 5.50s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key jic062-share-qlora-out-v7/config.json from bucket guanaco-mkml-models
Deleting key jic062-share-qlora-out-v7/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key jic062-share-qlora-out-v7/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key jic062-share-qlora-out-v7/tokenizer.json from bucket guanaco-mkml-models
Deleting key jic062-share-qlora-out-v7/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key jic062-share-qlora-out-v7_reward/config.json from bucket guanaco-reward-models
Deleting key jic062-share-qlora-out-v7_reward/merges.txt from bucket guanaco-reward-models
Deleting key jic062-share-qlora-out-v7_reward/reward.tensors from bucket guanaco-reward-models
Deleting key jic062-share-qlora-out-v7_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key jic062-share-qlora-out-v7_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key jic062-share-qlora-out-v7_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key jic062-share-qlora-out-v7_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.21s
jic062-share-qlora-out_v7 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics