Hugging Face's logo

tangledgroup
/
tangled-llama-33m-32k-base-v0.1

Text Generation
Transformers
Safetensors
litgpt
litdata
conversational
Inference Endpoints
Model card Files Community
Marko Tasic commited on
Commit
db520d1
•
2 Parent(s): 415ceb4 d8dedb3

Merge branch 'main' of hf.co:tangledgroup/tangled-llama-33m-32k-base-v0.1

Browse files
Files changed (1) hide show
  1. README.md +53 -79
README.md CHANGED
@@ -26,6 +26,10 @@ tags:
26
 
27
  ![logo](./misc/logo.png)
28
 
 
 
 
 
29
  [loss, val_loss](https://api.wandb.ai/links/mtasic85/z591qpyv)
30
 
31
  [val_ppl](https://api.wandb.ai/links/mtasic85/lku5hbws)
@@ -36,83 +40,53 @@ tags:
36
 
37
  ## lm-evaluation-harness
38
 
39
- | Tasks |Version| Filter |n-shot| Metric | |Value | |Stderr|
40
- |---------------------------------------|------:|----------------|-----:|-----------|---|-----:|---|-----:|
41
- |arc_challenge | 1|none | 0|acc |↑ |0.2065|± |0.0118|
42
- | | |none | 0|acc_norm |↑ |0.2338|± |0.0124|
43
- |gsm8k | 3|flexible-extract| 5|exact_match|↑ |0.0144|± |0.0033|
44
- | | |strict-match | 5|exact_match|↑ |0.0000|± |0.0000|
45
- |hellaswag | 1|none | 0|acc |↑ |0.2614|± |0.0044|
46
- | | |none | 0|acc_norm |↑ |0.2604|± |0.0044|
47
- |mmlu | 2|none | |acc |↑ |0.2280|± |0.0035|
48
- | - humanities | 2|none | |acc |↑ |0.2376|± |0.0062|
49
- | - formal_logic | 1|none | 0|acc |↑ |0.2698|± |0.0397|
50
- | - high_school_european_history | 1|none | 0|acc |↑ |0.2000|± |0.0312|
51
- | - high_school_us_history | 1|none | 0|acc |↑ |0.2500|± |0.0304|
52
- | - high_school_world_history | 1|none | 0|acc |↑ |0.2785|± |0.0292|
53
- | - international_law | 1|none | 0|acc |↑ |0.2149|± |0.0375|
54
- | - jurisprudence | 1|none | 0|acc |↑ |0.2407|± |0.0413|
55
- | - logical_fallacies | 1|none | 0|acc |↑ |0.2270|± |0.0329|
56
- | - moral_disputes | 1|none | 0|acc |↑ |0.2341|± |0.0228|
57
- | - moral_scenarios | 1|none | 0|acc |↑ |0.2380|± |0.0142|
58
- | - philosophy | 1|none | 0|acc |↑ |0.1833|± |0.0220|
59
- | - prehistory | 1|none | 0|acc |↑ |0.2160|± |0.0229|
60
- | - professional_law | 1|none | 0|acc |↑ |0.2399|± |0.0109|
61
- | - world_religions | 1|none | 0|acc |↑ |0.3275|± |0.0360|
62
- | - other | 2|none | |acc |↑ |0.2481|± |0.0077|
63
- | - business_ethics | 1|none | 0|acc |↑ |0.3100|± |0.0465|
64
- | - clinical_knowledge | 1|none | 0|acc |↑ |0.2264|± |0.0258|
65
- | - college_medicine | 1|none | 0|acc |↑ |0.2486|± |0.0330|
66
- | - global_facts | 1|none | 0|acc |↑ |0.1700|± |0.0378|
67
- | - human_aging | 1|none | 0|acc |↑ |0.3229|± |0.0314|
68
- | - management | 1|none | 0|acc |↑ |0.1845|± |0.0384|
69
- | - marketing | 1|none | 0|acc |↑ |0.3034|± |0.0301|
70
- | - medical_genetics | 1|none | 0|acc |↑ |0.3100|± |0.0465|
71
- | - miscellaneous | 1|none | 0|acc |↑ |0.2401|± |0.0153|
72
- | - nutrition | 1|none | 0|acc |↑ |0.2418|± |0.0245|
73
- | - professional_accounting | 1|none | 0|acc |↑ |0.2411|± |0.0255|
74
- | - professional_medicine | 1|none | 0|acc |↑ |0.1838|± |0.0235|
75
- | - virology | 1|none | 0|acc |↑ |0.2831|± |0.0351|
76
- | - social sciences | 2|none | |acc |↑ |0.2155|± |0.0074|
77
- | - econometrics | 1|none | 0|acc |↑ |0.2281|± |0.0395|
78
- | - high_school_geography | 1|none | 0|acc |↑ |0.1768|± |0.0272|
79
- | - high_school_government_and_politics| 1|none | 0|acc |↑ |0.1969|± |0.0287|
80
- | - high_school_macroeconomics | 1|none | 0|acc |↑ |0.2103|± |0.0207|
81
- | - high_school_microeconomics | 1|none | 0|acc |↑ |0.2143|± |0.0267|
82
- | - high_school_psychology | 1|none | 0|acc |↑ |0.1872|± |0.0167|
83
- | - human_sexuality | 1|none | 0|acc |↑ |0.2672|± |0.0388|
84
- | - professional_psychology | 1|none | 0|acc |↑ |0.2467|± |0.0174|
85
- | - public_relations | 1|none | 0|acc |↑ |0.2182|± |0.0396|
86
- | - security_studies | 1|none | 0|acc |↑ |0.1755|± |0.0244|
87
- | - sociology | 1|none | 0|acc |↑ |0.2438|± |0.0304|
88
- | - us_foreign_policy | 1|none | 0|acc |↑ |0.2700|± |0.0446|
89
- | - stem | 2|none | |acc |↑ |0.2058|± |0.0072|
90
- | - abstract_algebra | 1|none | 0|acc |↑ |0.1700|± |0.0378|
91
- | - anatomy | 1|none | 0|acc |↑ |0.1778|± |0.0330|
92
- | - astronomy | 1|none | 0|acc |↑ |0.1776|± |0.0311|
93
- | - college_biology | 1|none | 0|acc |↑ |0.2569|± |0.0365|
94
- | - college_chemistry | 1|none | 0|acc |↑ |0.1900|± |0.0394|
95
- | - college_computer_science | 1|none | 0|acc |↑ |0.2600|± |0.0441|
96
- | - college_mathematics | 1|none | 0|acc |↑ |0.2100|± |0.0409|
97
- | - college_physics | 1|none | 0|acc |↑ |0.2059|± |0.0402|
98
- | - computer_security | 1|none | 0|acc |↑ |0.2400|± |0.0429|
99
- | - conceptual_physics | 1|none | 0|acc |↑ |0.2681|± |0.0290|
100
- | - electrical_engineering | 1|none | 0|acc |↑ |0.2345|± |0.0353|
101
- | - elementary_mathematics | 1|none | 0|acc |↑ |0.2011|± |0.0206|
102
- | - high_school_biology | 1|none | 0|acc |↑ |0.1806|± |0.0219|
103
- | - high_school_chemistry | 1|none | 0|acc |↑ |0.1478|± |0.0250|
104
- | - high_school_computer_science | 1|none | 0|acc |↑ |0.2400|± |0.0429|
105
- | - high_school_mathematics | 1|none | 0|acc |↑ |0.2111|± |0.0249|
106
- | - high_school_physics | 1|none | 0|acc |↑ |0.1788|± |0.0313|
107
- | - high_school_statistics | 1|none | 0|acc |↑ |0.1620|± |0.0251|
108
- | - machine_learning | 1|none | 0|acc |↑ |0.2768|± |0.0425|
109
- |truthfulqa_mc2 | 2|none | 0|acc |↑ |0.4975|± |0.0165|
110
- |winogrande | 1|none | 0|acc |↑ |0.5146|± |0.0140|
111
 
112
- | Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
113
- |------------------|------:|------|------|------|---|-----:|---|-----:|
114
- |mmlu | 2|none | |acc |↑ |0.2280|± |0.0035|
115
- | - humanities | 2|none | |acc |↑ |0.2376|± |0.0062|
116
- | - other | 2|none | |acc |↑ |0.2481|± |0.0077|
117
- | - social sciences| 2|none | |acc |↑ |0.2155|± |0.0074|
118
- | - stem | 2|none | |acc |↑ |0.2058|± |0.0072|
 
26
 
27
  ![logo](./misc/logo.png)
28
 
29
+ A pretrained language model based on the Llama model with about **33M** parameters. This model has been trained on **9.7B** (`9,782,206,713`) tokens from more than **5.2M** (`5,285,575`) dataset rows.
30
+
31
+ This model is designed not for immediate use but rather for Continued Pretraining and Finetuning on a downstream task. While it can handle a context length of up to **32K** (`32,768`) tokens, it was pretrained with sequences of **2K** (`2048`) tokens.
32
+
33
  [loss, val_loss](https://api.wandb.ai/links/mtasic85/z591qpyv)
34
 
35
  [val_ppl](https://api.wandb.ai/links/mtasic85/lku5hbws)
 
40
 
41
  ## lm-evaluation-harness
42
 
43
+ | Tasks |Version|Filter|n-shot| Metric | |Value | |Stderr|
44
+ |-----------------------------------------------------------|-------|------|-----:|-----------------------|---|-----:|---|------|
45
+ |leaderboard | N/A| | | | | | | |
46
+ | - leaderboard_bbh | N/A| | | | | | | |
47
+ | - leaderboard_bbh_boolean_expressions | 1|none | 3|acc_norm |↑ |0.4600|± |0.0316|
48
+ | - leaderboard_bbh_causal_judgement | 1|none | 3|acc_norm |↑ |0.5187|± |0.0366|
49
+ | - leaderboard_bbh_date_understanding | 1|none | 3|acc_norm |↑ |0.1600|± |0.0232|
50
+ | - leaderboard_bbh_disambiguation_qa | 1|none | 3|acc_norm |↑ |0.3000|± |0.0290|
51
+ | - leaderboard_bbh_formal_fallacies | 1|none | 3|acc_norm |↑ |0.4680|± |0.0316|
52
+ | - leaderboard_bbh_geometric_shapes | 1|none | 3|acc_norm |↑ |0.0400|± |0.0124|
53
+ | - leaderboard_bbh_hyperbaton | 1|none | 3|acc_norm |↑ |0.5240|± |0.0316|
54
+ | - leaderboard_bbh_logical_deduction_five_objects | 1|none | 3|acc_norm |↑ |0.2240|± |0.0264|
55
+ | - leaderboard_bbh_logical_deduction_seven_objects | 1|none | 3|acc_norm |↑ |0.1480|± |0.0225|
56
+ | - leaderboard_bbh_logical_deduction_three_objects | 1|none | 3|acc_norm |↑ |0.3160|± |0.0295|
57
+ | - leaderboard_bbh_movie_recommendation | 1|none | 3|acc_norm |↑ |0.2560|± |0.0277|
58
+ | - leaderboard_bbh_navigate | 1|none | 3|acc_norm |↑ |0.4200|± |0.0313|
59
+ | - leaderboard_bbh_object_counting | 1|none | 3|acc_norm |↑ |0.0680|± |0.0160|
60
+ | - leaderboard_bbh_penguins_in_a_table | 1|none | 3|acc_norm |↑ |0.1918|± |0.0327|
61
+ | - leaderboard_bbh_reasoning_about_colored_objects | 1|none | 3|acc_norm |↑ |0.1160|± |0.0203|
62
+ | - leaderboard_bbh_ruin_names | 1|none | 3|acc_norm |↑ |0.2000|± |0.0253|
63
+ | - leaderboard_bbh_salient_translation_error_detection | 1|none | 3|acc_norm |↑ |0.1320|± |0.0215|
64
+ | - leaderboard_bbh_snarks | 1|none | 3|acc_norm |↑ |0.4719|± |0.0375|
65
+ | - leaderboard_bbh_sports_understanding | 1|none | 3|acc_norm |↑ |0.4600|± |0.0316|
66
+ | - leaderboard_bbh_temporal_sequences | 1|none | 3|acc_norm |↑ |0.2880|± |0.0287|
67
+ | - leaderboard_bbh_tracking_shuffled_objects_five_objects | 1|none | 3|acc_norm |↑ |0.1880|± |0.0248|
68
+ | - leaderboard_bbh_tracking_shuffled_objects_seven_objects| 1|none | 3|acc_norm |↑ |0.1280|± |0.0212|
69
+ | - leaderboard_bbh_tracking_shuffled_objects_three_objects| 1|none | 3|acc_norm |↑ |0.3040|± |0.0292|
70
+ | - leaderboard_bbh_web_of_lies | 1|none | 3|acc_norm |↑ |0.4880|± |0.0317|
71
+ | - leaderboard_gpqa | N/A| | | | | | | |
72
+ | - leaderboard_gpqa_diamond | 1|none | 0|acc_norm |↑ |0.2222|± |0.0296|
73
+ | - leaderboard_gpqa_extended | 1|none | 0|acc_norm |↑ |0.2381|± |0.0182|
74
+ | - leaderboard_gpqa_main | 1|none | 0|acc_norm |↑ |0.2478|± |0.0204|
75
+ | - leaderboard_ifeval | 3|none | 0|inst_level_loose_acc |↑ |0.1859|± | N/A|
76
+ | | |none | 0|inst_level_strict_acc |↑ |0.1811|± | N/A|
77
+ | | |none | 0|prompt_level_loose_acc |↑ |0.0906|± |0.0124|
78
+ | | |none | 0|prompt_level_strict_acc|↑ |0.0832|± |0.0119|
79
+ | - leaderboard_math_hard | N/A| | | | | | | |
80
+ | - leaderboard_math_algebra_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
81
+ | - leaderboard_math_counting_and_prob_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
82
+ | - leaderboard_math_geometry_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
83
+ | - leaderboard_math_intermediate_algebra_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
84
+ | - leaderboard_math_num_theory_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
85
+ | - leaderboard_math_prealgebra_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
86
+ | - leaderboard_math_precalculus_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
87
+ | - leaderboard_mmlu_pro | 0.1|none | 5|acc |↑ |0.1100|± |0.0029|
88
+ | - leaderboard_musr | N/A| | | | | | | |
89
+ | - leaderboard_musr_murder_mysteries | 1|none | 0|acc_norm |↑ |0.5000|± |0.0317|
90
+ | - leaderboard_musr_object_placements | 1|none | 0|acc_norm |↑ |0.2930|± |0.0285|
91
+ | - leaderboard_musr_team_allocation | 1|none | 0|acc_norm |↑ |0.3720|± |0.0306|
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
92