Update README.md
Browse files
README.md
CHANGED
|
@@ -1,11 +1,163 @@
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
-
|
|
|
|
| 4 |
---
|
| 5 |
|
| 6 |
-
# Model Card for
|
| 7 |
-
|
| 8 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
|
| 10 |
|
| 11 |
|
|
@@ -196,6 +348,4 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
|
|
| 196 |
|
| 197 |
## Model Card Contact
|
| 198 |
|
| 199 |
-
[More Information Needed]
|
| 200 |
-
|
| 201 |
-
|
|
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
---
|
| 6 |
|
| 7 |
+
# Model Card for nano-phi-v0.1
|
| 8 |
+
|
| 9 |
+
Inspired by [Phi2](https://huggingface.co/microsoft/phi-2), and open source small language model attempts like [smol_llama-101M-GQA](https://huggingface.co/BEE-spoke-data/smol_llama-101M-GQA).
|
| 10 |
+
Pre-trained with training 7B token from scratch, with a high quality dataset of 0.6B token.
|
| 11 |
+
It just took 2d 4h to train in Colab with a A100 40GB (~USD$ 100).
|
| 12 |
+
It achieves quite competitive results in evaluation given its training token, and training data size.
|
| 13 |
+
No alignment has been done yet.
|
| 14 |
+
|
| 15 |
+
hf-causal-experimental (pretrained=/content/lm-evaluation-harness/artifacts/checkpoint-pegfss6f:v13,use_accelerate=false,trust_remote_code=True), limit: None, provide_description: False, num_fewshot: 0, batch_size: 16
|
| 16 |
+
| Task |Version| Metric |Value | |Stderr|
|
| 17 |
+
|--------|------:|--------|-----:|---|-----:|
|
| 18 |
+
|arc_easy| 0|acc |0.4263|± |0.0101|
|
| 19 |
+
| | |acc_norm|0.3864|± |0.0100|
|
| 20 |
+
|
| 21 |
+
hf-causal-experimental (pretrained=/content/lm-evaluation-harness/artifacts/checkpoint-pegfss6f:v13,use_accelerate=false,trust_remote_code=True), limit: None, provide_description: False, num_fewshot: 25, batch_size: 16
|
| 22 |
+
| Task |Version| Metric |Value | |Stderr|
|
| 23 |
+
|-------------|------:|--------|-----:|---|-----:|
|
| 24 |
+
|arc_challenge| 0|acc |0.1826|± |0.0113|
|
| 25 |
+
| | |acc_norm|0.2193|± |0.0121|
|
| 26 |
+
|
| 27 |
+
hf-causal-experimental (pretrained=/content/lm-evaluation-harness/artifacts/checkpoint-pegfss6f:v13,use_accelerate=false,trust_remote_code=True), limit: None, provide_description: False, num_fewshot: 10, batch_size: 16
|
| 28 |
+
| Task |Version| Metric |Value | |Stderr|
|
| 29 |
+
|---------|------:|--------|-----:|---|-----:|
|
| 30 |
+
|hellaswag| 0|acc |0.2733|± |0.0044|
|
| 31 |
+
| | |acc_norm|0.2787|± |0.0045|
|
| 32 |
+
|
| 33 |
+
hf-causal-experimental (pretrained=/content/lm-evaluation-harness/artifacts/checkpoint-pegfss6f:v13,use_accelerate=false,trust_remote_code=True), limit: None, provide_description: False, num_fewshot: 0, batch_size: 16
|
| 34 |
+
| Task |Version|Metric|Value | |Stderr|
|
| 35 |
+
|-------------|------:|------|-----:|---|-----:|
|
| 36 |
+
|truthfulqa_mc| 1|mc1 |0.2521|± |0.0152|
|
| 37 |
+
| | |mc2 |0.4601|± |0.0154|
|
| 38 |
+
|
| 39 |
+
hf-causal-experimental (pretrained=/content/lm-evaluation-harness/artifacts/checkpoint-pegfss6f:v13,use_accelerate=false,trust_remote_code=True), limit: None, provide_description: False, num_fewshot: 5, batch_size: 16
|
| 40 |
+
| Task |Version| Metric |Value | |Stderr|
|
| 41 |
+
|-------------------------------------------------|------:|--------|-----:|---|-----:|
|
| 42 |
+
|hendrycksTest-abstract_algebra | 1|acc |0.2300|± |0.0423|
|
| 43 |
+
| | |acc_norm|0.2300|± |0.0423|
|
| 44 |
+
|hendrycksTest-anatomy | 1|acc |0.3111|± |0.0400|
|
| 45 |
+
| | |acc_norm|0.3111|± |0.0400|
|
| 46 |
+
|hendrycksTest-astronomy | 1|acc |0.2171|± |0.0336|
|
| 47 |
+
| | |acc_norm|0.2171|± |0.0336|
|
| 48 |
+
|hendrycksTest-business_ethics | 1|acc |0.2500|± |0.0435|
|
| 49 |
+
| | |acc_norm|0.2500|± |0.0435|
|
| 50 |
+
|hendrycksTest-clinical_knowledge | 1|acc |0.2226|± |0.0256|
|
| 51 |
+
| | |acc_norm|0.2226|± |0.0256|
|
| 52 |
+
|hendrycksTest-college_biology | 1|acc |0.2292|± |0.0351|
|
| 53 |
+
| | |acc_norm|0.2292|± |0.0351|
|
| 54 |
+
|hendrycksTest-college_chemistry | 1|acc |0.1700|± |0.0378|
|
| 55 |
+
| | |acc_norm|0.1700|± |0.0378|
|
| 56 |
+
|hendrycksTest-college_computer_science | 1|acc |0.2500|± |0.0435|
|
| 57 |
+
| | |acc_norm|0.2500|± |0.0435|
|
| 58 |
+
|hendrycksTest-college_mathematics | 1|acc |0.2500|± |0.0435|
|
| 59 |
+
| | |acc_norm|0.2500|± |0.0435|
|
| 60 |
+
|hendrycksTest-college_medicine | 1|acc |0.2023|± |0.0306|
|
| 61 |
+
| | |acc_norm|0.2023|± |0.0306|
|
| 62 |
+
|hendrycksTest-college_physics | 1|acc |0.3235|± |0.0466|
|
| 63 |
+
| | |acc_norm|0.3235|± |0.0466|
|
| 64 |
+
|hendrycksTest-computer_security | 1|acc |0.2600|± |0.0441|
|
| 65 |
+
| | |acc_norm|0.2600|± |0.0441|
|
| 66 |
+
|hendrycksTest-conceptual_physics | 1|acc |0.2511|± |0.0283|
|
| 67 |
+
| | |acc_norm|0.2511|± |0.0283|
|
| 68 |
+
|hendrycksTest-econometrics | 1|acc |0.2281|± |0.0395|
|
| 69 |
+
| | |acc_norm|0.2281|± |0.0395|
|
| 70 |
+
|hendrycksTest-electrical_engineering | 1|acc |0.2276|± |0.0349|
|
| 71 |
+
| | |acc_norm|0.2276|± |0.0349|
|
| 72 |
+
|hendrycksTest-elementary_mathematics | 1|acc |0.2460|± |0.0222|
|
| 73 |
+
| | |acc_norm|0.2460|± |0.0222|
|
| 74 |
+
|hendrycksTest-formal_logic | 1|acc |0.1508|± |0.0320|
|
| 75 |
+
| | |acc_norm|0.1508|± |0.0320|
|
| 76 |
+
|hendrycksTest-global_facts | 1|acc |0.3000|± |0.0461|
|
| 77 |
+
| | |acc_norm|0.3000|± |0.0461|
|
| 78 |
+
|hendrycksTest-high_school_biology | 1|acc |0.3387|± |0.0269|
|
| 79 |
+
| | |acc_norm|0.3387|± |0.0269|
|
| 80 |
+
|hendrycksTest-high_school_chemistry | 1|acc |0.2906|± |0.0319|
|
| 81 |
+
| | |acc_norm|0.2906|± |0.0319|
|
| 82 |
+
|hendrycksTest-high_school_computer_science | 1|acc |0.3100|± |0.0465|
|
| 83 |
+
| | |acc_norm|0.3100|± |0.0465|
|
| 84 |
+
|hendrycksTest-high_school_european_history | 1|acc |0.2182|± |0.0323|
|
| 85 |
+
| | |acc_norm|0.2182|± |0.0323|
|
| 86 |
+
|hendrycksTest-high_school_geography | 1|acc |0.3232|± |0.0333|
|
| 87 |
+
| | |acc_norm|0.3232|± |0.0333|
|
| 88 |
+
|hendrycksTest-high_school_government_and_politics| 1|acc |0.2021|± |0.0290|
|
| 89 |
+
| | |acc_norm|0.2021|± |0.0290|
|
| 90 |
+
|hendrycksTest-high_school_macroeconomics | 1|acc |0.2487|± |0.0219|
|
| 91 |
+
| | |acc_norm|0.2487|± |0.0219|
|
| 92 |
+
|hendrycksTest-high_school_mathematics | 1|acc |0.2741|± |0.0272|
|
| 93 |
+
| | |acc_norm|0.2741|± |0.0272|
|
| 94 |
+
|hendrycksTest-high_school_microeconomics | 1|acc |0.3319|± |0.0306|
|
| 95 |
+
| | |acc_norm|0.3319|± |0.0306|
|
| 96 |
+
|hendrycksTest-high_school_physics | 1|acc |0.3179|± |0.0380|
|
| 97 |
+
| | |acc_norm|0.3179|± |0.0380|
|
| 98 |
+
|hendrycksTest-high_school_psychology | 1|acc |0.2477|± |0.0185|
|
| 99 |
+
| | |acc_norm|0.2477|± |0.0185|
|
| 100 |
+
|hendrycksTest-high_school_statistics | 1|acc |0.4722|± |0.0340|
|
| 101 |
+
| | |acc_norm|0.4722|± |0.0340|
|
| 102 |
+
|hendrycksTest-high_school_us_history | 1|acc |0.2696|± |0.0311|
|
| 103 |
+
| | |acc_norm|0.2696|± |0.0311|
|
| 104 |
+
|hendrycksTest-high_school_world_history | 1|acc |0.2152|± |0.0268|
|
| 105 |
+
| | |acc_norm|0.2152|± |0.0268|
|
| 106 |
+
|hendrycksTest-human_aging | 1|acc |0.1973|± |0.0267|
|
| 107 |
+
| | |acc_norm|0.1973|± |0.0267|
|
| 108 |
+
|hendrycksTest-human_sexuality | 1|acc |0.2824|± |0.0395|
|
| 109 |
+
| | |acc_norm|0.2824|± |0.0395|
|
| 110 |
+
|hendrycksTest-international_law | 1|acc |0.2231|± |0.0380|
|
| 111 |
+
| | |acc_norm|0.2231|± |0.0380|
|
| 112 |
+
|hendrycksTest-jurisprudence | 1|acc |0.2222|± |0.0402|
|
| 113 |
+
| | |acc_norm|0.2222|± |0.0402|
|
| 114 |
+
|hendrycksTest-logical_fallacies | 1|acc |0.2822|± |0.0354|
|
| 115 |
+
| | |acc_norm|0.2822|± |0.0354|
|
| 116 |
+
|hendrycksTest-machine_learning | 1|acc |0.2768|± |0.0425|
|
| 117 |
+
| | |acc_norm|0.2768|± |0.0425|
|
| 118 |
+
|hendrycksTest-management | 1|acc |0.2039|± |0.0399|
|
| 119 |
+
| | |acc_norm|0.2039|± |0.0399|
|
| 120 |
+
|hendrycksTest-marketing | 1|acc |0.1966|± |0.0260|
|
| 121 |
+
| | |acc_norm|0.1966|± |0.0260|
|
| 122 |
+
|hendrycksTest-medical_genetics | 1|acc |0.2800|± |0.0451|
|
| 123 |
+
| | |acc_norm|0.2800|± |0.0451|
|
| 124 |
+
|hendrycksTest-miscellaneous | 1|acc |0.2746|± |0.0160|
|
| 125 |
+
| | |acc_norm|0.2746|± |0.0160|
|
| 126 |
+
|hendrycksTest-moral_disputes | 1|acc |0.2081|± |0.0219|
|
| 127 |
+
| | |acc_norm|0.2081|± |0.0219|
|
| 128 |
+
|hendrycksTest-moral_scenarios | 1|acc |0.2469|± |0.0144|
|
| 129 |
+
| | |acc_norm|0.2469|± |0.0144|
|
| 130 |
+
|hendrycksTest-nutrition | 1|acc |0.2647|± |0.0253|
|
| 131 |
+
| | |acc_norm|0.2647|± |0.0253|
|
| 132 |
+
|hendrycksTest-philosophy | 1|acc |0.1897|± |0.0223|
|
| 133 |
+
| | |acc_norm|0.1897|± |0.0223|
|
| 134 |
+
|hendrycksTest-prehistory | 1|acc |0.2377|± |0.0237|
|
| 135 |
+
| | |acc_norm|0.2377|± |0.0237|
|
| 136 |
+
|hendrycksTest-professional_accounting | 1|acc |0.2482|± |0.0258|
|
| 137 |
+
| | |acc_norm|0.2482|± |0.0258|
|
| 138 |
+
|hendrycksTest-professional_law | 1|acc |0.2464|± |0.0110|
|
| 139 |
+
| | |acc_norm|0.2464|± |0.0110|
|
| 140 |
+
|hendrycksTest-professional_medicine | 1|acc |0.4265|± |0.0300|
|
| 141 |
+
| | |acc_norm|0.4265|± |0.0300|
|
| 142 |
+
|hendrycksTest-professional_psychology | 1|acc |0.2614|± |0.0178|
|
| 143 |
+
| | |acc_norm|0.2614|± |0.0178|
|
| 144 |
+
|hendrycksTest-public_relations | 1|acc |0.1818|± |0.0369|
|
| 145 |
+
| | |acc_norm|0.1818|± |0.0369|
|
| 146 |
+
|hendrycksTest-security_studies | 1|acc |0.1959|± |0.0254|
|
| 147 |
+
| | |acc_norm|0.1959|± |0.0254|
|
| 148 |
+
|hendrycksTest-sociology | 1|acc |0.2289|± |0.0297|
|
| 149 |
+
| | |acc_norm|0.2289|± |0.0297|
|
| 150 |
+
|hendrycksTest-us_foreign_policy | 1|acc |0.2400|± |0.0429|
|
| 151 |
+
| | |acc_norm|0.2400|± |0.0429|
|
| 152 |
+
|hendrycksTest-virology | 1|acc |0.2048|± |0.0314|
|
| 153 |
+
| | |acc_norm|0.2048|± |0.0314|
|
| 154 |
+
|hendrycksTest-world_religions | 1|acc |0.2222|± |0.0319|
|
| 155 |
+
| | |acc_norm|0.2222|± |0.0319|
|
| 156 |
+
|
| 157 |
+
hf-causal-experimental (pretrained=/content/lm-evaluation-harness/artifacts/checkpoint-pegfss6f:v13,use_accelerate=false,trust_remote_code=True), limit: None, provide_description: False, num_fewshot: 5, batch_size: 16
|
| 158 |
+
| Task |Version|Metric|Value | |Stderr|
|
| 159 |
+
|----------|------:|------|-----:|---|-----:|
|
| 160 |
+
|winogrande| 0|acc |0.5099|± | 0.014|
|
| 161 |
|
| 162 |
|
| 163 |
|
|
|
|
| 348 |
|
| 349 |
## Model Card Contact
|
| 350 |
|
| 351 |
+
[More Information Needed]
|
|
|
|
|
|