submission_id: nousresearch-meta-llama_4941_v78
developer_uid: robert_irvine
status: inactive
model_repo: NousResearch/Meta-Llama-3-8B-Instruct
reward_repo: ChaiML/gpt2_medium_pairwise_60m_step_937500
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: 2024-07-03T18:28:11+00:00
model_name: nousresearch-meta-llama_4941_v78
model_group: NousResearch/Meta-Llama-
num_battles: 13904
num_wins: 7385
celo_rating: 1190.47
propriety_score: 0.7386834433021329
propriety_total_count: 6517.0
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_v78
ineligible_reason: None
language_model: NousResearch/Meta-Llama-3-8B-Instruct
model_size: 8B
reward_model: ChaiML/gpt2_medium_pairwise_60m_step_937500
us_pacific_date: 2024-07-03
win_ratio: 0.5311421173762946
Resubmit model
Running pipeline stage MKMLizer
Starting job with name nousresearch-meta-llama-4941-v78-mkmlizer
Waiting for job on nousresearch-meta-llama-4941-v78-mkmlizer to finish
nousresearch-meta-llama-4941-v78-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
nousresearch-meta-llama-4941-v78-mkmlizer: ║ _____ __ __ ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ /___/ ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ Version: 0.8.14 ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ https://mk1.ai ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ The license key for the current software has been verified as ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ belonging to: ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ Chai Research Corp. ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
nousresearch-meta-llama-4941-v78-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v78-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
nousresearch-meta-llama-4941-v78-mkmlizer: Downloaded to shared memory in 24.553s
nousresearch-meta-llama-4941-v78-mkmlizer: quantizing model to /dev/shm/model_cache
nousresearch-meta-llama-4941-v78-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
nousresearch-meta-llama-4941-v78-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 4%|▍ | 12/291 [00:00<00:02, 112.29it/s] Loading 0: 8%|▊ | 24/291 [00:00<00:02, 103.04it/s] Loading 0: 13%|█▎ | 39/291 [00:00<00:02, 116.05it/s] Loading 0: 18%|█▊ | 51/291 [00:00<00:02, 108.49it/s] Loading 0: 23%|██▎ | 66/291 [00:00<00:01, 116.91it/s] Loading 0: 27%|██▋ | 78/291 [00:00<00:02, 106.47it/s] Loading 0: 31%|███ | 89/291 [00:01<00:03, 56.56it/s] Loading 0: 35%|███▌ | 102/291 [00:01<00:02, 68.69it/s] Loading 0: 39%|███▉ | 113/291 [00:01<00:02, 73.93it/s] Loading 0: 43%|████▎ | 126/291 [00:01<00:01, 85.63it/s] Loading 0: 47%|████▋ | 137/291 [00:01<00:01, 90.06it/s] Loading 0: 51%|█████ | 148/291 [00:01<00:01, 83.41it/s] Loading 0: 54%|█████▍ | 158/291 [00:01<00:01, 81.99it/s] Loading 0: 59%|█████▉ | 173/291 [00:01<00:01, 95.70it/s] Loading 0: 64%|██████▍ | 187/291 [00:02<00:01, 63.07it/s] Loading 0: 67%|██████▋ | 196/291 [00:02<00:01, 62.75it/s] Loading 0: 72%|███████▏ | 210/291 [00:02<00:01, 75.70it/s] Loading 0: 76%|███████▌ | 220/291 [00:02<00:00, 78.04it/s] Loading 0: 79%|███████▊ | 229/291 [00:02<00:00, 80.03it/s] Loading 0: 83%|████████▎ | 241/291 [00:02<00:00, 89.55it/s] Loading 0: 88%|████████▊ | 255/291 [00:03<00:00, 99.76it/s] Loading 0: 91%|█████████▏| 266/291 [00:03<00:00, 91.06it/s] Loading 0: 95%|█████████▍| 276/291 [00:03<00:00, 92.02it/s] Loading 0: 99%|█████████▊| 287/291 [00:09<00:00, 6.08it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
nousresearch-meta-llama-4941-v78-mkmlizer: quantized model in 24.890s
nousresearch-meta-llama-4941-v78-mkmlizer: Processed model NousResearch/Meta-Llama-3-8B-Instruct in 49.443s
nousresearch-meta-llama-4941-v78-mkmlizer: creating bucket guanaco-mkml-models
nousresearch-meta-llama-4941-v78-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
nousresearch-meta-llama-4941-v78-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v78
nousresearch-meta-llama-4941-v78-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v78/config.json
nousresearch-meta-llama-4941-v78-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v78/tokenizer_config.json
nousresearch-meta-llama-4941-v78-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v78/special_tokens_map.json
nousresearch-meta-llama-4941-v78-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v78/tokenizer.json
nousresearch-meta-llama-4941-v78-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v78/flywheel_model.0.safetensors
nousresearch-meta-llama-4941-v78-mkmlizer: loading reward model from ChaiML/gpt2_medium_pairwise_60m_step_937500
nousresearch-meta-llama-4941-v78-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-v78-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v78-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-v78-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v78-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-v78-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v78-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-v78-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v78-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
nousresearch-meta-llama-4941-v78-mkmlizer: Saving duration: 0.440s
nousresearch-meta-llama-4941-v78-mkmlizer: Processed model ChaiML/gpt2_medium_pairwise_60m_step_937500 in 7.259s
nousresearch-meta-llama-4941-v78-mkmlizer: creating bucket guanaco-reward-models
nousresearch-meta-llama-4941-v78-mkmlizer: Bucket 's3://guanaco-reward-models/' created
nousresearch-meta-llama-4941-v78-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/nousresearch-meta-llama-4941-v78_reward
nousresearch-meta-llama-4941-v78-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v78_reward/config.json
nousresearch-meta-llama-4941-v78-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v78_reward/tokenizer_config.json
nousresearch-meta-llama-4941-v78-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v78_reward/special_tokens_map.json
nousresearch-meta-llama-4941-v78-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v78_reward/vocab.json
nousresearch-meta-llama-4941-v78-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/nousresearch-meta-llama-4941-v78_reward/merges.txt
nousresearch-meta-llama-4941-v78-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v78_reward/tokenizer.json
nousresearch-meta-llama-4941-v78-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/nousresearch-meta-llama-4941-v78_reward/reward.tensors
Job nousresearch-meta-llama-4941-v78-mkmlizer completed after 84.06s with status: succeeded
Stopping job with name nousresearch-meta-llama-4941-v78-mkmlizer
Pipeline stage MKMLizer completed in 84.84s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service nousresearch-meta-llama-4941-v78
Waiting for inference service nousresearch-meta-llama-4941-v78 to be ready
Failed to get response for submission neversleep-noromaid-v0_8068_v109: ('http://neversleep-noromaid-v0-8068-v109-predictor-default.tenant-chaiml-guanaco.knative.ord1.coreweave.cloud/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:51178->127.0.0.1:8080: read: connection reset by peer\n')
Inference service nousresearch-meta-llama-4941-v78 ready after 60.37396740913391s
Pipeline stage ISVCDeployer completed in 66.95s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2782394886016846s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Received healthy response to inference request in 1.2629077434539795s
Received healthy response to inference request in 1.239952564239502s
Received healthy response to inference request in 1.2782618999481201s
Received healthy response to inference request in 1.4360933303833008s
5 requests
0 failed requests
5th percentile: 1.2445436000823975
10th percentile: 1.249134635925293
20th percentile: 1.258316707611084
30th percentile: 1.2659785747528076
40th percentile: 1.2721202373504639
50th percentile: 1.2782618999481201
60th percentile: 1.3413944721221924
70th percentile: 1.4045270442962645
80th percentile: 1.6045225620269776
90th percentile: 1.941381025314331
95th percentile: 2.1098102569580077
99th percentile: 2.2445536422729493
mean time: 1.4990910053253175
Pipeline stage StressChecker completed in 8.18s
nousresearch-meta-llama_4941_v78 status is now deployed due to DeploymentManager action
nousresearch-meta-llama_4941_v78 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics