submission_id: nousresearch-meta-llama_4941_v73
developer_uid: robert_irvine
status: inactive
model_repo: NousResearch/Meta-Llama-3-8B-Instruct
reward_repo: rirv938/reward_gpt2_medium_preference_24m_e2
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': ['</s>', '<|user|>', '###', '\n'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
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}
reward_formatter: {'memory_template': 'Memory: {memory}\n', 'prompt_template': '{prompt}\n', 'bot_template': 'Bot: {message}\n', 'user_template': 'User: {message}\n', 'response_template': 'Bot:', 'truncate_by_message': False}
timestamp: 2024-07-03T18:22:28+00:00
model_name: nousresearch-meta-llama_4941_v73
model_group: NousResearch/Meta-Llama-
num_battles: 14108
num_wins: 7304
celo_rating: 1181.74
propriety_score: 0.7335594219994102
propriety_total_count: 6782.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: nousresearch-meta-llama_4941_v73
ineligible_reason: None
language_model: NousResearch/Meta-Llama-3-8B-Instruct
model_size: 8B
reward_model: rirv938/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-07-03
win_ratio: 0.5177204423022399
Resubmit model
Running pipeline stage MKMLizer
Starting job with name nousresearch-meta-llama-4941-v73-mkmlizer
Waiting for job on nousresearch-meta-llama-4941-v73-mkmlizer to finish
nousresearch-meta-llama-4941-v73-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
nousresearch-meta-llama-4941-v73-mkmlizer: ║ _____ __ __ ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ /___/ ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ Version: 0.8.14 ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ https://mk1.ai ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ The license key for the current software has been verified as ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ belonging to: ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ Chai Research Corp. ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
nousresearch-meta-llama-4941-v73-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v73-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
nousresearch-meta-llama-4941-v73-mkmlizer: Downloaded to shared memory in 27.081s
nousresearch-meta-llama-4941-v73-mkmlizer: quantizing model to /dev/shm/model_cache
nousresearch-meta-llama-4941-v73-mkmlizer: Saving flywheel model at /dev/shm/model_cache
nousresearch-meta-llama-4941-v73-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:05, 48.00it/s] Loading 0: 6%|▌ | 17/291 [00:00<00:03, 89.48it/s] Loading 0: 10%|█ | 30/291 [00:00<00:02, 99.78it/s] Loading 0: 14%|█▎ | 40/291 [00:00<00:02, 92.71it/s] Loading 0: 17%|█▋ | 50/291 [00:00<00:02, 90.35it/s] Loading 0: 22%|██▏ | 63/291 [00:00<00:02, 100.99it/s] Loading 0: 26%|██▌ | 75/291 [00:00<00:02, 103.53it/s] Loading 0: 30%|██▉ | 86/291 [00:01<00:03, 53.77it/s] Loading 0: 34%|███▍ | 100/291 [00:01<00:02, 69.04it/s] Loading 0: 38%|███▊ | 111/291 [00:01<00:02, 75.12it/s] Loading 0: 42%|████▏ | 121/291 [00:01<00:02, 78.21it/s] Loading 0: 45%|████▌ | 131/291 [00:01<00:02, 79.76it/s] Loading 0: 49%|████▉ | 143/291 [00:01<00:01, 88.76it/s] Loading 0: 54%|█████▎ | 156/291 [00:01<00:01, 95.24it/s] Loading 0: 57%|█████▋ | 167/291 [00:01<00:01, 91.93it/s] Loading 0: 62%|██████▏ | 181/291 [00:02<00:01, 98.92it/s] Loading 0: 66%|██████▌ | 192/291 [00:02<00:01, 56.17it/s] Loading 0: 69%|██████▉ | 202/291 [00:02<00:01, 61.81it/s] Loading 0: 73%|███████▎ | 212/291 [00:02<00:01, 66.91it/s] Loading 0: 77%|███████▋ | 224/291 [00:02<00:00, 77.49it/s] Loading 0: 80%|████████ | 234/291 [00:02<00:00, 80.04it/s] Loading 0: 85%|████████▍ | 246/291 [00:03<00:00, 86.31it/s] Loading 0: 88%|████████▊ | 256/291 [00:03<00:00, 83.90it/s] Loading 0: 91%|█████████▏| 266/291 [00:03<00:00, 82.82it/s] Loading 0: 96%|█████████▌| 278/291 [00:03<00:00, 91.42it/s] Loading 0: 99%|█████████▉| 288/291 [00:09<00:00, 5.85it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
nousresearch-meta-llama-4941-v73-mkmlizer: quantized model in 30.345s
nousresearch-meta-llama-4941-v73-mkmlizer: Processed model NousResearch/Meta-Llama-3-8B-Instruct in 57.427s
nousresearch-meta-llama-4941-v73-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
nousresearch-meta-llama-4941-v73-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v73
nousresearch-meta-llama-4941-v73-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v73/config.json
nousresearch-meta-llama-4941-v73-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v73/tokenizer_config.json
nousresearch-meta-llama-4941-v73-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v73/special_tokens_map.json
nousresearch-meta-llama-4941-v73-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v73/tokenizer.json
nousresearch-meta-llama-4941-v73-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v73/flywheel_model.0.safetensors
nousresearch-meta-llama-4941-v73-mkmlizer: loading reward model from rirv938/reward_gpt2_medium_preference_24m_e2
nousresearch-meta-llama-4941-v73-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.
nousresearch-meta-llama-4941-v73-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v73-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`.
nousresearch-meta-llama-4941-v73-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v73-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.
nousresearch-meta-llama-4941-v73-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v73-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.
nousresearch-meta-llama-4941-v73-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v73-mkmlizer: /opt/conda/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
nousresearch-meta-llama-4941-v73-mkmlizer: return self.fget.__get__(instance, owner)()
nousresearch-meta-llama-4941-v73-mkmlizer: Saving duration: 0.417s
nousresearch-meta-llama-4941-v73-mkmlizer: Processed model rirv938/reward_gpt2_medium_preference_24m_e2 in 4.125s
nousresearch-meta-llama-4941-v73-mkmlizer: creating bucket guanaco-reward-models
nousresearch-meta-llama-4941-v73-mkmlizer: Bucket 's3://guanaco-reward-models/' created
nousresearch-meta-llama-4941-v73-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/nousresearch-meta-llama-4941-v73_reward
nousresearch-meta-llama-4941-v73-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v73_reward/tokenizer_config.json
nousresearch-meta-llama-4941-v73-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v73_reward/config.json
nousresearch-meta-llama-4941-v73-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v73_reward/special_tokens_map.json
nousresearch-meta-llama-4941-v73-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v73_reward/vocab.json
nousresearch-meta-llama-4941-v73-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/nousresearch-meta-llama-4941-v73_reward/merges.txt
nousresearch-meta-llama-4941-v73-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v73_reward/tokenizer.json
nousresearch-meta-llama-4941-v73-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/nousresearch-meta-llama-4941-v73_reward/reward.tensors
Job nousresearch-meta-llama-4941-v73-mkmlizer completed after 83.8s with status: succeeded
Stopping job with name nousresearch-meta-llama-4941-v73-mkmlizer
Pipeline stage MKMLizer completed in 84.65s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.12s
Running pipeline stage ISVCDeployer
Creating inference service nousresearch-meta-llama-4941-v73
Waiting for inference service nousresearch-meta-llama-4941-v73 to be ready
Inference service nousresearch-meta-llama-4941-v73 ready after 40.19084191322327s
Pipeline stage ISVCDeployer completed in 46.89s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.10022234916687s
Received healthy response to inference request in 1.2369866371154785s
Received healthy response to inference request in 1.2373456954956055s
Received healthy response to inference request in 1.2414705753326416s
Received healthy response to inference request in 1.2191636562347412s
5 requests
0 failed requests
5th percentile: 1.2227282524108887
10th percentile: 1.2262928485870361
20th percentile: 1.233422040939331
30th percentile: 1.237058448791504
40th percentile: 1.2372020721435546
50th percentile: 1.2373456954956055
60th percentile: 1.23899564743042
70th percentile: 1.2406455993652343
80th percentile: 1.4132209300994876
90th percentile: 1.7567216396331788
95th percentile: 1.9284719944000241
99th percentile: 2.065872278213501
mean time: 1.4070377826690674
Pipeline stage StressChecker completed in 7.83s
nousresearch-meta-llama_4941_v73 status is now deployed due to DeploymentManager action
nousresearch-meta-llama_4941_v73 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics