submission_id: function_tifib_2024-08-15
developer_uid: end_to_end_test
alignment_samples: 6836
alignment_score: 2.120002693077629
celo_rating: 1145.79
display_name: gpt4-tl
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}
generation_params: {'temperature': 1.5, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 100, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', 'You:'], 'max_input_tokens': 512, 'best_of': 4, 'max_output_tokens': 64, 'reward_max_token_input': 256}
is_internal_developer: True
model_group:
model_name: gpt4-tl
num_battles: 6836
num_wins: 2775
propriety_score: 0.8223684210526315
propriety_total_count: 608.0
ranking_group: single
reward_repo: ChaiML/gpt2_xl_pairwise_89m_step_347634
status: torndown
submission_type: function
timestamp: 2024-08-15T01:25:44+00:00
us_pacific_date: 2024-08-14
win_ratio: 0.40593914569923933
Download Preferencedata
Resubmit model
Running pipeline stage StressChecker
HTTP Request: %s %s "%s %d %s"
Received healthy response to inference request in 1.3100168704986572s
HTTP Request: %s %s "%s %d %s"
Received healthy response to inference request in 1.7503585815429688s
HTTP Request: %s %s "%s %d %s"
Received healthy response to inference request in 1.0181868076324463s
HTTP Request: %s %s "%s %d %s"
Received healthy response to inference request in 1.4557867050170898s
HTTP Request: %s %s "%s %d %s"
Received healthy response to inference request in 1.0507307052612305s
5 requests
0 failed requests
5th percentile: 1.024695587158203
10th percentile: 1.0312043666839599
20th percentile: 1.0442219257354737
30th percentile: 1.1025879383087158
40th percentile: 1.2063024044036865
50th percentile: 1.3100168704986572
60th percentile: 1.3683248043060303
70th percentile: 1.4266327381134032
80th percentile: 1.5147010803222656
90th percentile: 1.6325298309326173
95th percentile: 1.691444206237793
99th percentile: 1.7385757064819336
mean time: 1.3170159339904786
Pipeline stage StressChecker completed in 9.73s
function_tifib_2024-08-15 status is now deployed due to DeploymentManager action
function_tifib_2024-08-15 status is now inactive due to admin request
function_tifib_2024-08-15 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics