developer_uid: Hastagaras
submission_id: hastagaras-anjir-8b-l3_v10
model_name: t
model_group: Hastagaras/Anjir-8B-L3
status: torndown
timestamp: 2024-06-15T01:49:12+00:00
num_battles: 28281
num_wins: 15173
celo_rating: 1214.97
family_friendly_score: 0.0
submission_type: basic
model_repo: Hastagaras/Anjir-8B-L3
model_architecture: LlamaForCausalLM
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: t
is_internal_developer: False
language_model: Hastagaras/Anjir-8B-L3
model_size: 8B
ranking_group: single
us_pacific_date: 2024-06-14
win_ratio: 0.536508610020862
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.08, 'top_k': 100, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], '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}
model_eval_status: success
reward_formatter: {'bot_template': '{bot_name}: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '{user_name}: {message}\n'}
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-anjir-8b-l3-v10-mkmlizer
Waiting for job on hastagaras-anjir-8b-l3-v10-mkmlizer to finish
hastagaras-anjir-8b-l3-v10-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ _____ __ __ ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ /___/ ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ https://mk1.ai ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ belonging to: ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ Chai Research Corp. ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ║ ║
hastagaras-anjir-8b-l3-v10-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-anjir-8b-l3-v10-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/utils/_deprecation.py:131: FutureWarning: 'list_files_info' (from 'huggingface_hub.hf_api') is deprecated and will be removed from version '0.23'. Use `list_repo_tree` and `get_paths_info` instead.
hastagaras-anjir-8b-l3-v10-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-anjir-8b-l3-v10-mkmlizer: Downloaded to shared memory in 16.811s
hastagaras-anjir-8b-l3-v10-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-anjir-8b-l3-v10-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-anjir-8b-l3-v10-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<10:44, 2.23s/it] Loading 0: 5%|▌ | 15/291 [00:04<01:01, 4.46it/s] Loading 0: 11%|█ | 32/291 [00:04<00:22, 11.46it/s] Loading 0: 17%|█▋ | 50/291 [00:04<00:11, 20.90it/s] Loading 0: 22%|██▏ | 64/291 [00:05<00:09, 24.52it/s] Loading 0: 27%|██▋ | 78/291 [00:05<00:06, 33.54it/s] Loading 0: 33%|███▎ | 95/291 [00:05<00:04, 46.74it/s] Loading 0: 37%|███▋ | 109/291 [00:05<00:03, 58.22it/s] Loading 0: 42%|████▏ | 122/291 [00:05<00:02, 68.38it/s] Loading 0: 48%|████▊ | 139/291 [00:05<00:01, 84.06it/s] Loading 0: 53%|█████▎ | 153/291 [00:05<00:01, 93.06it/s] Loading 0: 57%|█████▋ | 166/291 [00:06<00:01, 65.85it/s] Loading 0: 63%|██████▎ | 184/291 [00:06<00:01, 83.06it/s] Loading 0: 69%|██████▉ | 202/291 [00:06<00:00, 99.26it/s] Loading 0: 76%|███████▌ | 220/291 [00:06<00:00, 113.85it/s] Loading 0: 82%|████████▏ | 238/291 [00:06<00:00, 125.29it/s] Loading 0: 88%|████████▊ | 256/291 [00:06<00:00, 135.52it/s] Loading 0: 93%|█████████▎| 272/291 [00:07<00:00, 87.53it/s] Loading 0: 99%|█████████▉| 288/291 [00:07<00:00, 99.90it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-anjir-8b-l3-v10-mkmlizer: quantized model in 23.091s
hastagaras-anjir-8b-l3-v10-mkmlizer: Processed model Hastagaras/Anjir-8B-L3 in 42.535s
hastagaras-anjir-8b-l3-v10-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-anjir-8b-l3-v10-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-anjir-8b-l3-v10-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-anjir-8b-l3-v10
hastagaras-anjir-8b-l3-v10-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-anjir-8b-l3-v10/config.json
hastagaras-anjir-8b-l3-v10-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-anjir-8b-l3-v10/special_tokens_map.json
hastagaras-anjir-8b-l3-v10-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-anjir-8b-l3-v10/tokenizer_config.json
hastagaras-anjir-8b-l3-v10-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-anjir-8b-l3-v10/tokenizer.json
hastagaras-anjir-8b-l3-v10-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-anjir-8b-l3-v10/flywheel_model.0.safetensors
hastagaras-anjir-8b-l3-v10-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-anjir-8b-l3-v10-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:913: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
hastagaras-anjir-8b-l3-v10-mkmlizer: warnings.warn(
hastagaras-anjir-8b-l3-v10-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:757: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
hastagaras-anjir-8b-l3-v10-mkmlizer: warnings.warn(
hastagaras-anjir-8b-l3-v10-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.
hastagaras-anjir-8b-l3-v10-mkmlizer: warnings.warn(
hastagaras-anjir-8b-l3-v10-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()
hastagaras-anjir-8b-l3-v10-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-anjir-8b-l3-v10-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-anjir-8b-l3-v10-mkmlizer: Saving duration: 0.398s
hastagaras-anjir-8b-l3-v10-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.287s
hastagaras-anjir-8b-l3-v10-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-anjir-8b-l3-v10_reward/reward.tensors
Job hastagaras-anjir-8b-l3-v10-mkmlizer completed after 72.8s with status: succeeded
Stopping job with name hastagaras-anjir-8b-l3-v10-mkmlizer
Pipeline stage MKMLizer completed in 77.12s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.08s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-anjir-8b-l3-v10
Waiting for inference service hastagaras-anjir-8b-l3-v10 to be ready
Inference service hastagaras-anjir-8b-l3-v10 ready after 40.2529935836792s
Pipeline stage ISVCDeployer completed in 47.65s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.069383382797241s
Received healthy response to inference request in 1.2989997863769531s
Received healthy response to inference request in 1.2970609664916992s
Received healthy response to inference request in 1.2789065837860107s
Received healthy response to inference request in 1.3441922664642334s
5 requests
0 failed requests
5th percentile: 1.2825374603271484
10th percentile: 1.2861683368682861
20th percentile: 1.2934300899505615
30th percentile: 1.29744873046875
40th percentile: 1.2982242584228516
50th percentile: 1.2989997863769531
60th percentile: 1.3170767784118653
70th percentile: 1.3351537704467773
80th percentile: 1.489230489730835
90th percentile: 1.7793069362640381
95th percentile: 1.9243451595306396
99th percentile: 2.0403757381439207
mean time: 1.4577085971832275
Pipeline stage StressChecker completed in 7.87s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running M-Eval for topic stay_in_character
Running pipeline stage DaemonicSafetyScorer
M-Eval Dataset for topic stay_in_character is loaded
Pipeline stage DaemonicSafetyScorer completed in 0.06s
hastagaras-anjir-8b-l3_v10 status is now deployed due to DeploymentManager action
hastagaras-anjir-8b-l3_v10 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of hastagaras-anjir-8b-l3_v10
Running pipeline stage ISVCDeleter
Checking if service hastagaras-anjir-8b-l3-v10 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 3.71s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key hastagaras-anjir-8b-l3-v10/config.json from bucket guanaco-mkml-models
Deleting key hastagaras-anjir-8b-l3-v10/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key hastagaras-anjir-8b-l3-v10/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key hastagaras-anjir-8b-l3-v10/tokenizer.json from bucket guanaco-mkml-models
Deleting key hastagaras-anjir-8b-l3-v10/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key hastagaras-anjir-8b-l3-v10_reward/config.json from bucket guanaco-reward-models
Deleting key hastagaras-anjir-8b-l3-v10_reward/merges.txt from bucket guanaco-reward-models
Deleting key hastagaras-anjir-8b-l3-v10_reward/reward.tensors from bucket guanaco-reward-models
Deleting key hastagaras-anjir-8b-l3-v10_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key hastagaras-anjir-8b-l3-v10_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key hastagaras-anjir-8b-l3-v10_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key hastagaras-anjir-8b-l3-v10_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.98s
hastagaras-anjir-8b-l3_v10 status is now torndown due to DeploymentManager action