Given-MPT-7B

This is a merge of the following MPT-7B models:

  • gorilla-llm/gorilla-mpt-7b-hf-v0
  • ibm/mpt-7b-instruct2
  • TehVenom/MPT-7b-WizardLM_Uncensored-Storywriter-Merge
  • emozilla/mpt-7b-storysummarizer
  • nomic-ai/gpt4all-mpt

Model License

Apache 2.0

Purpose

This model is for experimenting with merging and routing to expert layers.

Test eval on only 10% of eval set

hf-causal (pretrained=Multi-Domain-Expert-Layers/given-mpt-7b,dtype=bfloat16,trust_remote_code=True), limit: 0.1, provide_description: False, num_fewshot: 0, batch_size: None

Task Version Metric Value Stderr
arc_challenge 0 acc 0.4274 ± 0.0459
acc_norm 0.3846 ± 0.0452
arc_easy 0 acc 0.7863 ± 0.0381
acc_norm 0.7350 ± 0.0410
hellaswag 0 acc 0.5556 ± 0.0461
acc_norm 0.8120 ± 0.0363
hendrycksTest-college_chemistry 0 acc 0.3600 ± 0.0482
acc_norm 0.3700 ± 0.0485
hendrycksTest-college_computer_science 0 acc 0.3400 ± 0.0476
acc_norm 0.3600 ± 0.0482
hendrycksTest-college_mathematics 0 acc 0.2500 ± 0.0435
acc_norm 0.2900 ± 0.0456
hendrycksTest-college_medicine 0 acc 0.3675 ± 0.0448
acc_norm 0.3162 ± 0.0432
hendrycksTest-college_physics 0 acc 0.2451 ± 0.0428
acc_norm 0.2941 ± 0.0453
hendrycksTest-computer_security 0 acc 0.4800 ± 0.0502
acc_norm 0.4400 ± 0.0499
hendrycksTest-conceptual_physics 0 acc 0.2051 ± 0.0375
acc_norm 0.1709 ± 0.0350
hendrycksTest-econometrics 0 acc 0.2982 ± 0.0430
acc_norm 0.2368 ± 0.0400
hendrycksTest-electrical_engineering 0 acc 0.3248 ± 0.0435
acc_norm 0.3590 ± 0.0445
hendrycksTest-elementary_mathematics 0 acc 0.3333 ± 0.0438
acc_norm 0.3162 ± 0.0432
hendrycksTest-formal_logic 0 acc 0.3077 ± 0.0429
acc_norm 0.3248 ± 0.0435
hendrycksTest-global_facts 0 acc 0.3000 ± 0.0461
acc_norm 0.2700 ± 0.0446
hendrycksTest-high_school_biology 0 acc 0.3675 ± 0.0448
acc_norm 0.3077 ± 0.0429
hendrycksTest-high_school_chemistry 0 acc 0.2564 ± 0.0405
acc_norm 0.2906 ± 0.0422
hendrycksTest-high_school_computer_science 0 acc 0.4100 ± 0.0494
acc_norm 0.4400 ± 0.0499
hendrycksTest-high_school_european_history 0 acc 0.4359 ± 0.0460
acc_norm 0.3590 ± 0.0445
hendrycksTest-high_school_geography 0 acc 0.3248 ± 0.0435
acc_norm 0.3675 ± 0.0448
hendrycksTest-high_school_government_and_politics 0 acc 0.3932 ± 0.0454
acc_norm 0.3932 ± 0.0454
hendrycksTest-high_school_macroeconomics 0 acc 0.3333 ± 0.0438
acc_norm 0.3248 ± 0.0435
hendrycksTest-high_school_mathematics 0 acc 0.2051 ± 0.0375
acc_norm 0.2564 ± 0.0405
hendrycksTest-high_school_microeconomics 0 acc 0.3504 ± 0.0443
acc_norm 0.4188 ± 0.0458
hendrycksTest-high_school_physics 0 acc 0.2650 ± 0.0410
acc_norm 0.2906 ± 0.0422
hendrycksTest-high_school_psychology 0 acc 0.3761 ± 0.0450
acc_norm 0.3419 ± 0.0440
hendrycksTest-high_school_statistics 0 acc 0.3077 ± 0.0429
acc_norm 0.3504 ± 0.0443
hendrycksTest-high_school_us_history 0 acc 0.3333 ± 0.0438
acc_norm 0.3333 ± 0.0438
hendrycksTest-high_school_world_history 0 acc 0.3333 ± 0.0438
acc_norm 0.3419 ± 0.0440
hendrycksTest-human_aging 0 acc 0.3761 ± 0.0450
acc_norm 0.3162 ± 0.0432
hendrycksTest-human_sexuality 0 acc 0.4274 ± 0.0459
acc_norm 0.3761 ± 0.0450
hendrycksTest-international_law 0 acc 0.4188 ± 0.0458
acc_norm 0.4957 ± 0.0464
hendrycksTest-jurisprudence 0 acc 0.3148 ± 0.0449
acc_norm 0.4815 ± 0.0483
hendrycksTest-logical_fallacies 0 acc 0.3504 ± 0.0443
acc_norm 0.3675 ± 0.0448
hendrycksTest-machine_learning 0 acc 0.3214 ± 0.0443
acc_norm 0.2946 ± 0.0433
hendrycksTest-management 0 acc 0.3786 ± 0.0480
acc_norm 0.3495 ± 0.0472
hendrycksTest-marketing 0 acc 0.5043 ± 0.0464
acc_norm 0.4188 ± 0.0458
hendrycksTest-medical_genetics 0 acc 0.3200 ± 0.0469
acc_norm 0.4100 ± 0.0494
hendrycksTest-miscellaneous 0 acc 0.5299 ± 0.0463
acc_norm 0.4872 ± 0.0464
hendrycksTest-moral_disputes 0 acc 0.3248 ± 0.0435
acc_norm 0.3162 ± 0.0432
hendrycksTest-moral_scenarios 0 acc 0.3248 ± 0.0435
acc_norm 0.2479 ± 0.0401
hendrycksTest-nutrition 0 acc 0.3675 ± 0.0448
acc_norm 0.3932 ± 0.0454
hendrycksTest-philosophy 0 acc 0.2991 ± 0.0425
acc_norm 0.3504 ± 0.0443
hendrycksTest-prehistory 0 acc 0.2821 ± 0.0418
acc_norm 0.3248 ± 0.0435
hendrycksTest-professional_accounting 0 acc 0.2137 ± 0.0381
acc_norm 0.2222 ± 0.0386
hendrycksTest-professional_law 0 acc 0.3077 ± 0.0429
acc_norm 0.2735 ± 0.0414
hendrycksTest-professional_medicine 0 acc 0.2991 ± 0.0425
acc_norm 0.2650 ± 0.0410
hendrycksTest-professional_psychology 0 acc 0.3248 ± 0.0435
acc_norm 0.3419 ± 0.0440
hendrycksTest-public_relations 0 acc 0.3909 ± 0.0467
acc_norm 0.3545 ± 0.0458
hendrycksTest-security_studies 0 acc 0.3419 ± 0.0440
acc_norm 0.2906 ± 0.0422
hendrycksTest-sociology 0 acc 0.3761 ± 0.0450
acc_norm 0.3162 ± 0.0432
hendrycksTest-us_foreign_policy 0 acc 0.5000 ± 0.0503
acc_norm 0.4100 ± 0.0494
hendrycksTest-virology 0 acc 0.3932 ± 0.0454
acc_norm 0.3248 ± 0.0435
hendrycksTest-world_religions 0 acc 0.5299 ± 0.0463
acc_norm 0.5128 ± 0.0464
truthfulqa_gen 1 bleurt_max -0.8551 ± 0.0501
bleurt_acc 0.3590 ± 0.0445
bleurt_diff -0.1292 ± 0.0483
bleu_max 19.3738 ± 1.8461
bleu_acc 0.3932 ± 0.0454
bleu_diff -4.3883 ± 2.1748
rouge1_max 41.8428 ± 2.6156
rouge1_acc 0.3162 ± 0.0432
rouge1_diff -8.8583 ± 2.7745
rouge2_max 26.3956 ± 2.8311
rouge2_acc 0.2137 ± 0.0381
rouge2_diff -9.5287 ± 3.3258
rougeL_max 39.5215 ± 2.5620
rougeL_acc 0.3162 ± 0.0432
rougeL_diff -8.5753 ± 2.8259
Downloads last month
16
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model authors have turned it off explicitly.