diff --git a/README.md b/README.md new file mode 100644 index 0000000000000000000000000000000000000000..9311a4cf1cc71ccce0cdd7931ce71b99a9718981 --- /dev/null +++ b/README.md @@ -0,0 +1,86 @@ +--- +language: en +thumbnail: +license: mit +tags: +- question-answering +- bert +- bert-base +datasets: +- squad +metrics: +- squad +widget: +- text: "Where is located the Eiffel Tower?" + context: "The Eiffel Tower is a wrought-iron lattice tower on the Champ de Mars in Paris, France. It is named after the engineer Gustave Eiffel, whose company designed and built the tower." +- text: "Who is Frederic Chopin?" + context: "Frédéric François Chopin, born Fryderyk Franciszek Chopin (1 March 1810 – 17 October 1849), was a Polish composer and virtuoso pianist of the Romantic era who wrote primarily for solo piano." +--- + +## BERT-base uncased model fine-tuned on SQuAD v1 + +This model is [block-sparse](https://github.com/huggingface/pytorch_block_sparse). + +That means that with the right runtime it can run roughly 3x faster than an dense network, with 25% of the original weights. + +This of course has some impact on the accuracy (see below). + +It uses a modified version of Victor Sanh [Movement Pruning](https://arxiv.org/abs/2005.07683) method. + +This model was fine-tuned from the HuggingFace[BERT](https://www.aclweb.org/anthology/N19-1423/) base uncased checkpoint on [SQuAD1.1](https://rajpurkar.github.io/SQuAD-explorer), and distilled from the equivalent model [csarron/bert-base-uncased-squad-v1](https://huggingface.co/csarron/bert-base-uncased-squad-v1). +This model is case-insensitive: it does not make a difference between english and English. + +## Pruning details +![Pruning details](https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/pruning.svg) + +## Density plot + + + +## Details + +| Dataset | Split | # samples | +| -------- | ----- | --------- | +| SQuAD1.1 | train | 90.6K | +| SQuAD1.1 | eval | 11.1k | + +### Fine-tuning +- Python: `3.8.5` + +- Machine specs: + + `CPU: Intel(R) Core(TM) i7-6700K CPU` + + `Memory: 64 GiB` + + `GPUs: 1 GeForce GTX 3090, with 24GiB memory` + + `GPU driver: 455.23.05, CUDA: 11.1` + + +### Results + +**Pytorch model file size**: `342`M (original BERT:`438M`) + +| Metric | # Value | # Original ([Table 2](https://www.aclweb.org/anthology/N19-1423.pdf))| +| ------ | --------- | --------- | +| **EM** | **74.39** | **80.8** | +| **F1** | **83.26** | **88.5** | + +## Example Usage + +```python +from transformers import pipeline + +qa_pipeline = pipeline( + "question-answering", + model="madlag/bert-base-uncased-squad-v1-sparse0.25", + tokenizer="madlag/bert-base-uncased-squad-v1-sparse0.25" +) + +predictions = qa_pipeline({ + 'context': "Frédéric François Chopin, born Fryderyk Franciszek Chopin (1 March 1810 – 17 October 1849), was a Polish composer and virtuoso pianist of the Romantic era who wrote primarily for solo piano.", + 'question': "Who is Frederic Chopin?", +}) + +print(predictions) \ No newline at end of file diff --git a/model_card/density.js b/model_card/density.js new file mode 100644 index 0000000000000000000000000000000000000000..6ecdcc1dd2e241837f2d86c12e775b2fa157c211 --- /dev/null +++ b/model_card/density.js @@ -0,0 +1,174 @@ +(function() { + var fn = function() { + + (function(root) { + function now() { + return new Date(); + } + + var force = false; + + if (typeof root._bokeh_onload_callbacks === "undefined" || force === true) { + root._bokeh_onload_callbacks = []; + root._bokeh_is_loading = undefined; + } + + + + + var element = document.getElementById("4af67900-bd92-4310-973e-0cf20443efff"); + if (element == null) { + console.warn("Bokeh: autoload.js configured with elementid '4af67900-bd92-4310-973e-0cf20443efff' but no matching script tag was found.") + } + + + function run_callbacks() { + try { + root._bokeh_onload_callbacks.forEach(function(callback) { + if (callback != null) + callback(); + }); + } finally { + delete root._bokeh_onload_callbacks + } + console.debug("Bokeh: all callbacks have finished"); + } + + function load_libs(css_urls, js_urls, callback) { + if (css_urls == null) css_urls = []; + if (js_urls == null) js_urls = []; + + root._bokeh_onload_callbacks.push(callback); + if (root._bokeh_is_loading > 0) { + console.debug("Bokeh: BokehJS is being loaded, scheduling callback at", now()); + return null; + } + if (js_urls == null || js_urls.length === 0) { + run_callbacks(); + return null; + } + console.debug("Bokeh: BokehJS not loaded, scheduling load and callback at", now()); + root._bokeh_is_loading = css_urls.length + js_urls.length; + + function on_load() { + root._bokeh_is_loading--; + if (root._bokeh_is_loading === 0) { + console.debug("Bokeh: all BokehJS libraries/stylesheets loaded"); + run_callbacks() + } + } + + function on_error() { + console.error("failed to load " + url); + } + + for (var i = 0; i < css_urls.length; i++) { + var url = css_urls[i]; + const element = document.createElement("link"); + element.onload = on_load; + element.onerror = on_error; + element.rel = "stylesheet"; + element.type = "text/css"; + element.href = url; + console.debug("Bokeh: injecting link tag for BokehJS stylesheet: ", url); + document.body.appendChild(element); + } + + const hashes = {"https://cdn.bokeh.org/bokeh/release/bokeh-2.2.3.min.js": "T2yuo9Oe71Cz/I4X9Ac5+gpEa5a8PpJCDlqKYO0CfAuEszu1JrXLl8YugMqYe3sM", "https://cdn.bokeh.org/bokeh/release/bokeh-widgets-2.2.3.min.js": "98GDGJ0kOMCUMUePhksaQ/GYgB3+NH9h996V88sh3aOiUNX3N+fLXAtry6xctSZ6", "https://cdn.bokeh.org/bokeh/release/bokeh-tables-2.2.3.min.js": "89bArO+nlbP3sgakeHjCo1JYxYR5wufVgA3IbUvDY+K7w4zyxJqssu7wVnfeKCq8"}; + + for (var i = 0; i < js_urls.length; i++) { + var url = js_urls[i]; + var element = document.createElement('script'); + element.onload = on_load; + element.onerror = on_error; + element.async = false; + element.src = url; + if (url in hashes) { + element.crossOrigin = "anonymous"; + element.integrity = "sha384-" + hashes[url]; + } + console.debug("Bokeh: injecting script tag for BokehJS library: ", url); + document.head.appendChild(element); + } + }; + + function inject_raw_css(css) { + const element = document.createElement("style"); + element.appendChild(document.createTextNode(css)); + document.body.appendChild(element); + } + + + var js_urls = ["https://cdn.bokeh.org/bokeh/release/bokeh-2.2.3.min.js", "https://cdn.bokeh.org/bokeh/release/bokeh-widgets-2.2.3.min.js", "https://cdn.bokeh.org/bokeh/release/bokeh-tables-2.2.3.min.js"]; + var css_urls = []; + + + var inline_js = [ + function(Bokeh) { + Bokeh.set_log_level("info"); + }, + + function(Bokeh) { + (function() { + var fn = function() { + Bokeh.safely(function() { + (function(root) { + function embed_document(root) { + + var docs_json = '{"913c3228-db80-48c9-8309-97f33ce0691f":{"roots":{"references":[{"attributes":{"data_source":{"id":"1023"},"glyph":{"id":"1025"},"hover_glyph":null,"muted_glyph":null,"name":"query","nonselection_glyph":{"id":"1026"},"selection_glyph":null,"view":{"id":"1028"}},"id":"1027","type":"GlyphRenderer"},{"attributes":{"axis_label":"Density","formatter":{"id":"1031"},"minor_tick_line_color":null,"ticker":{"id":"1018"}},"id":"1017","type":"LinearAxis"},{"attributes":{},"id":"1051","type":"UnionRenderers"},{"attributes":{},"id":"1011","type":"LinearScale"},{"attributes":{},"id":"1052","type":"Selection"},{"attributes":{},"id":"1090","type":"Selection"},{"attributes":{"label":{"value":"value"},"renderers":[{"id":"1058"}]},"id":"1071","type":"LegendItem"},{"attributes":{},"id":"1069","type":"UnionRenderers"},{"attributes":{},"id":"1009","type":"LinearScale"},{"attributes":{},"id":"1033","type":"BasicTickFormatter"},{"attributes":{},"id":"1070","type":"Selection"},{"attributes":{"axis":{"id":"1017"},"dimension":1,"ticker":null},"id":"1020","type":"Grid"},{"attributes":{"items":[{"id":"1037"},{"id":"1053"},{"id":"1071"},{"id":"1091"}],"location":"top_left","orientation":"horizontal"},"id":"1036","type":"Legend"},{"attributes":{"fill_color":{"value":"#ed5642"},"line_color":{"value":"#ed5642"},"top":{"field":"height"},"width":{"value":0.125},"x":{"field":"x"}},"id":"1040","type":"VBar"},{"attributes":{"fill_alpha":{"value":0.1},"fill_color":{"value":"#6573f7"},"line_alpha":{"value":0.1},"line_color":{"value":"#6573f7"},"top":{"field":"height"},"width":{"value":0.125},"x":{"field":"x"}},"id":"1026","type":"VBar"},{"attributes":{"source":{"id":"1054"}},"id":"1059","type":"CDSView"},{"attributes":{"source":{"id":"1023"}},"id":"1028","type":"CDSView"},{"attributes":{"fill_color":{"value":"#6573f7"},"line_color":{"value":"#6573f7"},"top":{"field":"height"},"width":{"value":0.125},"x":{"field":"x"}},"id":"1025","type":"VBar"},{"attributes":{"label":{"value":"query"},"renderers":[{"id":"1027"}]},"id":"1037","type":"LegendItem"},{"attributes":{"data":{"density":["0.20","0.10","0.16","0.21","0.11","0.12","0.11","0.15","0.06","0.04","0.02","0.02"],"height":[0.1979166716337204,0.1024305522441864,0.1597222238779068,0.2083333283662796,0.1145833358168602,0.1180555522441864,0.1076388880610466,0.1475694477558136,0.0590277761220932,0.0364583320915699,0.0243055559694767,0.02083333395421505],"img_height":["96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px"],"img_width":["96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px"],"name":["0.attention.value","1.attention.value","2.attention.value","3.attention.value","4.attention.value","5.attention.value","6.attention.value","7.attention.value","8.attention.value","9.attention.value","10.attention.value","11.attention.value"],"url":["https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_0_attention_self_value.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_1_attention_self_value.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_2_attention_self_value.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_3_attention_self_value.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_4_attention_self_value.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_5_attention_self_value.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_6_attention_self_value.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_7_attention_self_value.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_8_attention_self_value.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_9_attention_self_value.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_10_attention_self_value.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_11_attention_self_value.png"],"x":[0.41666666666666663,1.4166666666666665,2.416666666666667,3.416666666666667,4.416666666666666,5.416666666666666,6.416666666666666,7.416666666666666,8.416666666666668,9.416666666666668,10.416666666666668,11.416666666666668]},"selected":{"id":"1070"},"selection_policy":{"id":"1069"}},"id":"1054","type":"ColumnDataSource"},{"attributes":{"fill_alpha":{"value":0.1},"fill_color":{"value":"#20cb97"},"line_alpha":{"value":0.1},"line_color":{"value":"#20cb97"},"top":{"field":"height"},"width":{"value":0.125},"x":{"field":"x"}},"id":"1057","type":"VBar"},{"attributes":{},"id":"1005","type":"DataRange1d"},{"attributes":{"data_source":{"id":"1054"},"glyph":{"id":"1056"},"hover_glyph":null,"muted_glyph":null,"name":"value","nonselection_glyph":{"id":"1057"},"selection_glyph":null,"view":{"id":"1059"}},"id":"1058","type":"GlyphRenderer"},{"attributes":{},"id":"1014","type":"BasicTicker"},{"attributes":{},"id":"1034","type":"UnionRenderers"},{"attributes":{"axis":{"id":"1013"},"grid_line_color":null,"ticker":null},"id":"1016","type":"Grid"},{"attributes":{"data":{"density":["0.15","0.13","0.17","0.23","0.11","0.10","0.09","0.11","0.07","0.12","0.06","0.06"],"height":[0.1527777761220932,0.1319444477558136,0.1684027761220932,0.2256944477558136,0.1111111119389534,0.0989583358168602,0.0920138880610466,0.1145833358168602,0.0694444477558136,0.1180555522441864,0.0642361119389534,0.0590277761220932],"img_height":["96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px"],"img_width":["96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px"],"name":["0.attention.query","1.attention.query","2.attention.query","3.attention.query","4.attention.query","5.attention.query","6.attention.query","7.attention.query","8.attention.query","9.attention.query","10.attention.query","11.attention.query"],"url":["https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_0_attention_self_query.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_1_attention_self_query.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_2_attention_self_query.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_3_attention_self_query.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_4_attention_self_query.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_5_attention_self_query.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_6_attention_self_query.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_7_attention_self_query.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_8_attention_self_query.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_9_attention_self_query.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_10_attention_self_query.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_11_attention_self_query.png"],"x":[0.08333333333333333,1.0833333333333333,2.0833333333333335,3.0833333333333335,4.083333333333333,5.083333333333333,6.083333333333333,7.083333333333333,8.083333333333334,9.083333333333334,10.083333333333334,11.083333333333334]},"selected":{"id":"1035"},"selection_policy":{"id":"1034"}},"id":"1023","type":"ColumnDataSource"},{"attributes":{},"id":"1035","type":"Selection"},{"attributes":{"data":{"density":["0.21","0.29","0.25","0.09","0.17","0.15","0.16","0.29","0.23","0.21","0.21","0.16","0.11","0.22","0.16","0.12","0.18","0.13","0.10","0.16","0.13","0.12","0.11","0.08","0.05","0.03","0.03","0.03","0.01","0.01","0.03","0.03","0.02","0.02","0.05","0.03"],"height":[0.2083333283662796,0.2894965410232544,0.2543402910232544,0.0885416641831398,0.1736111044883728,0.1488715261220932,0.1649305522441864,0.2938368022441864,0.2330729216337204,0.2083333283662796,0.2078993022441864,0.1584201455116272,0.1111111119389534,0.2239583283662796,0.1605902761220932,0.1197916641831398,0.1779513955116272,0.1267361044883728,0.0972222238779068,0.1627604216337204,0.1302083283662796,0.1197916641831398,0.1063368022441864,0.078125,0.0520833320915699,0.0342881940305233,0.02734375,0.0329861119389534,0.01128472201526165,0.0086805559694767,0.0295138880610466,0.0329861119389534,0.0234375,0.02083333395421505,0.0455729179084301,0.0290798619389534],"img_height":["96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px"],"img_width":["96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px","96px","96px","384px"],"name":["0.attention.output","0.intermediate","0.output","1.attention.output","1.intermediate","1.output","2.attention.output","2.intermediate","2.output","3.attention.output","3.intermediate","3.output","4.attention.output","4.intermediate","4.output","5.attention.output","5.intermediate","5.output","6.attention.output","6.intermediate","6.output","7.attention.output","7.intermediate","7.output","8.attention.output","8.intermediate","8.output","9.attention.output","9.intermediate","9.output","10.attention.output","10.intermediate","10.output","11.attention.output","11.intermediate","11.output"],"url":["https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_0_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_0_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_0_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_1_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_1_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_1_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_2_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_2_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_2_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_3_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_3_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_3_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_4_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_4_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_4_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_5_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_5_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_5_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_6_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_6_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_6_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_7_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_7_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_7_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_8_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_8_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_8_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_9_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_9_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_9_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_10_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_10_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_10_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_11_attention_output_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_11_intermediate_dense.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_11_output_dense.png"],"x":[0.5833333333333334,0.75,0.9166666666666667,1.5833333333333333,1.75,1.9166666666666665,2.5833333333333335,2.75,2.916666666666667,3.5833333333333335,3.75,3.916666666666667,4.583333333333333,4.75,4.916666666666666,5.583333333333333,5.75,5.916666666666666,6.583333333333333,6.75,6.916666666666666,7.583333333333333,7.75,7.916666666666666,8.583333333333334,8.75,8.916666666666668,9.583333333333334,9.75,9.916666666666668,10.583333333333334,10.75,10.916666666666668,11.583333333333334,11.75,11.916666666666668]},"selected":{"id":"1090"},"selection_policy":{"id":"1089"}},"id":"1072","type":"ColumnDataSource"},{"attributes":{},"id":"1031","type":"BasicTickFormatter"},{"attributes":{"label":{"value":"fully connected"},"renderers":[{"id":"1076"}]},"id":"1091","type":"LegendItem"},{"attributes":{},"id":"1018","type":"BasicTicker"},{"attributes":{"text":"Transformer Layers"},"id":"1003","type":"Title"},{"attributes":{"fill_color":{"value":"#20cb97"},"line_color":{"value":"#20cb97"},"top":{"field":"height"},"width":{"value":0.125},"x":{"field":"x"}},"id":"1056","type":"VBar"},{"attributes":{"below":[{"id":"1013"}],"center":[{"id":"1016"},{"id":"1020"},{"id":"1036"}],"left":[{"id":"1017"}],"outline_line_color":null,"plot_height":300,"plot_width":505,"renderers":[{"id":"1027"},{"id":"1042"},{"id":"1058"},{"id":"1076"}],"title":{"id":"1003"},"toolbar":{"id":"1021"},"x_range":{"id":"1005"},"x_scale":{"id":"1009"},"y_range":{"id":"1007"},"y_scale":{"id":"1011"}},"id":"1002","subtype":"Figure","type":"Plot"},{"attributes":{"data":{"density":["0.16","0.14","0.17","0.25","0.12","0.11","0.11","0.13","0.08","0.11","0.06","0.06"],"height":[0.1597222238779068,0.1423611044883728,0.1684027761220932,0.2517361044883728,0.1180555522441864,0.1128472238779068,0.1059027761220932,0.1319444477558136,0.0833333358168602,0.109375,0.0590277761220932,0.0607638880610466],"img_height":["96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px"],"img_width":["96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px","96px"],"name":["0.attention.key","1.attention.key","2.attention.key","3.attention.key","4.attention.key","5.attention.key","6.attention.key","7.attention.key","8.attention.key","9.attention.key","10.attention.key","11.attention.key"],"url":["https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_0_attention_self_key.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_1_attention_self_key.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_2_attention_self_key.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_3_attention_self_key.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_4_attention_self_key.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_5_attention_self_key.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_6_attention_self_key.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_7_attention_self_key.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_8_attention_self_key.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_9_attention_self_key.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_10_attention_self_key.png","https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.13-v1/raw/main/model_card/layer_images/layer_11_attention_self_key.png"],"x":[0.25,1.25,2.25,3.25,4.25,5.25,6.25,7.25,8.25,9.25,10.25,11.25]},"selected":{"id":"1052"},"selection_policy":{"id":"1051"}},"id":"1038","type":"ColumnDataSource"},{"attributes":{"label":{"value":"key"},"renderers":[{"id":"1042"}]},"id":"1053","type":"LegendItem"},{"attributes":{"fill_color":{"value":"#aa69f7"},"line_color":{"value":"#aa69f7"},"top":{"field":"height"},"width":{"value":0.125},"x":{"field":"x"}},"id":"1074","type":"VBar"},{"attributes":{"data_source":{"id":"1072"},"glyph":{"id":"1074"},"hover_glyph":null,"muted_glyph":null,"name":"fully connected","nonselection_glyph":{"id":"1075"},"selection_glyph":null,"view":{"id":"1077"}},"id":"1076","type":"GlyphRenderer"},{"attributes":{"data_source":{"id":"1038"},"glyph":{"id":"1040"},"hover_glyph":null,"muted_glyph":null,"name":"key","nonselection_glyph":{"id":"1041"},"selection_glyph":null,"view":{"id":"1043"}},"id":"1042","type":"GlyphRenderer"},{"attributes":{"fill_alpha":{"value":0.1},"fill_color":{"value":"#aa69f7"},"line_alpha":{"value":0.1},"line_color":{"value":"#aa69f7"},"top":{"field":"height"},"width":{"value":0.125},"x":{"field":"x"}},"id":"1075","type":"VBar"},{"attributes":{"fill_alpha":{"value":0.1},"fill_color":{"value":"#ed5642"},"line_alpha":{"value":0.1},"line_color":{"value":"#ed5642"},"top":{"field":"height"},"width":{"value":0.125},"x":{"field":"x"}},"id":"1041","type":"VBar"},{"attributes":{"active_drag":"auto","active_inspect":"auto","active_multi":null,"active_scroll":"auto","active_tap":"auto","tools":[{"id":"1001"}]},"id":"1021","type":"Toolbar"},{"attributes":{"source":{"id":"1072"}},"id":"1077","type":"CDSView"},{"attributes":{"axis_label":"Layer","formatter":{"id":"1033"},"minor_tick_line_color":null,"ticker":{"id":"1014"}},"id":"1013","type":"LinearAxis"},{"attributes":{"source":{"id":"1038"}},"id":"1043","type":"CDSView"},{"attributes":{"callback":null,"tooltips":"\\n <div>\\n <div style=\\"margin-bottom:10px\\">\\n <span style=\\"font-size: 15px;\\"><b>@name</b><br/>density=@density</span>\\n </div>\\n <div> \\n <img\\n src=\\"@url\\" height=\\"@img_height\\" width=\\"@img_width\\" alt=\\"@url\\"\\n style=\\"float: left; margin: 0px 15px 15px 0px;\\"\\n border=\\"0\\"\\n />\\n </div>\\n </div>\\n "},"id":"1001","type":"HoverTool"},{"attributes":{"start":0},"id":"1007","type":"DataRange1d"},{"attributes":{},"id":"1089","type":"UnionRenderers"}],"root_ids":["1002"]},"title":"Bokeh Application","version":"2.2.3"}}'; + var render_items = [{"docid":"913c3228-db80-48c9-8309-97f33ce0691f","root_ids":["1002"],"roots":{"1002":"4af67900-bd92-4310-973e-0cf20443efff"}}]; + root.Bokeh.embed.embed_items(docs_json, render_items); + + } + if (root.Bokeh !== undefined) { + embed_document(root); + } else { + var attempts = 0; + var timer = setInterval(function(root) { + if (root.Bokeh !== undefined) { + clearInterval(timer); + embed_document(root); + } else { + attempts++; + if (attempts > 100) { + clearInterval(timer); + console.log("Bokeh: ERROR: Unable to run BokehJS code because BokehJS library is missing"); + } + } + }, 10, root) + } + })(window); + }); + }; + if (document.readyState != "loading") fn(); + else document.addEventListener("DOMContentLoaded", fn); + })(); + }, + function(Bokeh) { + + + } + ]; + + function run_inline_js() { + + for (var i = 0; i < inline_js.length; i++) { + inline_js[i].call(root, root.Bokeh); + } + + } + + if (root._bokeh_is_loading === 0) { + console.debug("Bokeh: BokehJS loaded, going straight to plotting"); + run_inline_js(); + } else { + load_libs(css_urls, js_urls, function() { + console.debug("Bokeh: BokehJS plotting callback run at", now()); + run_inline_js(); + }); + } + }(window)); + }; + if (document.readyState != "loading") fn(); + else document.addEventListener("DOMContentLoaded", fn); +})(); \ No newline at end of file diff --git a/model_card/layer_images/layer_0_attention_output_dense.png b/model_card/layer_images/layer_0_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..d4652c3c6ef1bcd25d982ee749506dad84357448 Binary files /dev/null and b/model_card/layer_images/layer_0_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_0_attention_self_key.png b/model_card/layer_images/layer_0_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..5bdbdeeef3b048a8f2adfc0631f59c9a54dd261a Binary files /dev/null and b/model_card/layer_images/layer_0_attention_self_key.png differ diff --git a/model_card/layer_images/layer_0_attention_self_query.png b/model_card/layer_images/layer_0_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..526fcce4dd6be082dcad87b63a17c0399c1004fc Binary files /dev/null and b/model_card/layer_images/layer_0_attention_self_query.png differ diff --git a/model_card/layer_images/layer_0_attention_self_value.png b/model_card/layer_images/layer_0_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..e15c081f46e55302e6bd71e9ea6bc755a34add59 Binary files /dev/null and b/model_card/layer_images/layer_0_attention_self_value.png differ diff --git a/model_card/layer_images/layer_0_intermediate_dense.png b/model_card/layer_images/layer_0_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..2242d1a9d0a6f2dd465c1bbbbd35f5dd89ea6ad3 Binary files /dev/null and b/model_card/layer_images/layer_0_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_0_output_dense.png b/model_card/layer_images/layer_0_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..7d0735aa5b5199d2217f31545091738d2eb0927f Binary files /dev/null and b/model_card/layer_images/layer_0_output_dense.png differ diff --git a/model_card/layer_images/layer_10_attention_output_dense.png b/model_card/layer_images/layer_10_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..a9103f741167a58b282d53cd7e0ccc0fb26a4a89 Binary files /dev/null and b/model_card/layer_images/layer_10_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_10_attention_self_key.png b/model_card/layer_images/layer_10_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..78632b5f90ca865250e9cab8e91f5e910af20e61 Binary files /dev/null and b/model_card/layer_images/layer_10_attention_self_key.png differ diff --git a/model_card/layer_images/layer_10_attention_self_query.png b/model_card/layer_images/layer_10_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..f23ddb916263dfce3d9d42e7445275b9f33c9b71 Binary files /dev/null and b/model_card/layer_images/layer_10_attention_self_query.png differ diff --git a/model_card/layer_images/layer_10_attention_self_value.png b/model_card/layer_images/layer_10_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..d570e312e37df289b6825717f02827e6c205ff61 Binary files /dev/null and b/model_card/layer_images/layer_10_attention_self_value.png differ diff --git a/model_card/layer_images/layer_10_intermediate_dense.png b/model_card/layer_images/layer_10_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..313f4fe99829e6ea36a7875c251786a2a9272694 Binary files /dev/null and b/model_card/layer_images/layer_10_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_10_output_dense.png b/model_card/layer_images/layer_10_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..16cfa9560440d29b18012208dab43e85d2775819 Binary files /dev/null and b/model_card/layer_images/layer_10_output_dense.png differ diff --git a/model_card/layer_images/layer_11_attention_output_dense.png b/model_card/layer_images/layer_11_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..8b5270723069cd7ca90578daa9ec477676ee738d Binary files /dev/null and b/model_card/layer_images/layer_11_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_11_attention_self_key.png b/model_card/layer_images/layer_11_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..ea2aa464fdb50988371c633c588d16f58fb5c7f9 Binary files /dev/null and b/model_card/layer_images/layer_11_attention_self_key.png differ diff --git a/model_card/layer_images/layer_11_attention_self_query.png b/model_card/layer_images/layer_11_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..8e5765056f4ae4cfa035707c2418d903ad0dcdec Binary files /dev/null and b/model_card/layer_images/layer_11_attention_self_query.png differ diff --git a/model_card/layer_images/layer_11_attention_self_value.png b/model_card/layer_images/layer_11_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..6f24776efe41bd96adbf74e673a0b888bcd5c648 Binary files /dev/null and b/model_card/layer_images/layer_11_attention_self_value.png differ diff --git a/model_card/layer_images/layer_11_intermediate_dense.png b/model_card/layer_images/layer_11_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..8c8480f6fda7115ce093113c5e7926223e42af40 Binary files /dev/null and b/model_card/layer_images/layer_11_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_11_output_dense.png b/model_card/layer_images/layer_11_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..8aa62bdf0aed51acf64ac4df4a9f5c933cd3247e Binary files /dev/null and b/model_card/layer_images/layer_11_output_dense.png differ diff --git a/model_card/layer_images/layer_1_attention_output_dense.png b/model_card/layer_images/layer_1_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..4e53aa54b15c3209d9ee8f3c4193b6311401e704 Binary files /dev/null and b/model_card/layer_images/layer_1_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_1_attention_self_key.png b/model_card/layer_images/layer_1_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..b3c157180708a2df3741749495ce339111ad469b Binary files /dev/null and b/model_card/layer_images/layer_1_attention_self_key.png differ diff --git a/model_card/layer_images/layer_1_attention_self_query.png b/model_card/layer_images/layer_1_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..97447d6cbd184f3d2b477910048f63d0f7dbf6b0 Binary files /dev/null and b/model_card/layer_images/layer_1_attention_self_query.png differ diff --git a/model_card/layer_images/layer_1_attention_self_value.png b/model_card/layer_images/layer_1_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..1689e7e23195a95de054b9a496fa436c9a1861bc Binary files /dev/null and b/model_card/layer_images/layer_1_attention_self_value.png differ diff --git a/model_card/layer_images/layer_1_intermediate_dense.png b/model_card/layer_images/layer_1_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..400f5c1b8287962f8bd6e840c36bcc7bd43ccb4e Binary files /dev/null and b/model_card/layer_images/layer_1_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_1_output_dense.png b/model_card/layer_images/layer_1_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..bd49ef3d9972096b0f477584b9d298ee4f5e7fd2 Binary files /dev/null and b/model_card/layer_images/layer_1_output_dense.png differ diff --git a/model_card/layer_images/layer_2_attention_output_dense.png b/model_card/layer_images/layer_2_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..74fdd29abc84bc33c114cd966afe8f2ec9ac802b Binary files /dev/null and b/model_card/layer_images/layer_2_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_2_attention_self_key.png b/model_card/layer_images/layer_2_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..1ac58415463701ac58e806bc75ca6ee3d78b43a4 Binary files /dev/null and b/model_card/layer_images/layer_2_attention_self_key.png differ diff --git a/model_card/layer_images/layer_2_attention_self_query.png b/model_card/layer_images/layer_2_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..a66bd3309bc17e78819c12b2c2107d6de07acb89 Binary files /dev/null and b/model_card/layer_images/layer_2_attention_self_query.png differ diff --git a/model_card/layer_images/layer_2_attention_self_value.png b/model_card/layer_images/layer_2_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..9f933d3641a173a857c42bdfefe7d69ff7b80732 Binary files /dev/null and b/model_card/layer_images/layer_2_attention_self_value.png differ diff --git a/model_card/layer_images/layer_2_intermediate_dense.png b/model_card/layer_images/layer_2_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..0bb92074415f89583bbfec806d7b83aaafa847a4 Binary files /dev/null and b/model_card/layer_images/layer_2_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_2_output_dense.png b/model_card/layer_images/layer_2_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..8b88d33f907e9fc26273fb16097dae21104de6db Binary files /dev/null and b/model_card/layer_images/layer_2_output_dense.png differ diff --git a/model_card/layer_images/layer_3_attention_output_dense.png b/model_card/layer_images/layer_3_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..f8175b696ee9950b9e7cd7fdb8a236115d0d49cc Binary files /dev/null and b/model_card/layer_images/layer_3_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_3_attention_self_key.png b/model_card/layer_images/layer_3_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..c77a043065828beea647c228445e3052105879b4 Binary files /dev/null and b/model_card/layer_images/layer_3_attention_self_key.png differ diff --git a/model_card/layer_images/layer_3_attention_self_query.png b/model_card/layer_images/layer_3_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..8f0dd6b156109641abba9f783f7a784bb6894d91 Binary files /dev/null and b/model_card/layer_images/layer_3_attention_self_query.png differ diff --git a/model_card/layer_images/layer_3_attention_self_value.png b/model_card/layer_images/layer_3_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..7f8854293f6d65c699fd656d7f1d57ec0f9287ba Binary files /dev/null and b/model_card/layer_images/layer_3_attention_self_value.png differ diff --git a/model_card/layer_images/layer_3_intermediate_dense.png b/model_card/layer_images/layer_3_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..d45cce5074098a7186d88c02c32264b2c5463d3c Binary files /dev/null and b/model_card/layer_images/layer_3_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_3_output_dense.png b/model_card/layer_images/layer_3_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..389c98b753cdb92c0e8d43c312c6afe8eec04ba0 Binary files /dev/null and b/model_card/layer_images/layer_3_output_dense.png differ diff --git a/model_card/layer_images/layer_4_attention_output_dense.png b/model_card/layer_images/layer_4_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..06a63ff5273a17b274afcb9c36b591bce1353144 Binary files /dev/null and b/model_card/layer_images/layer_4_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_4_attention_self_key.png b/model_card/layer_images/layer_4_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..1ed081bdd9fa68efe77e279785dd76170cec64bb Binary files /dev/null and b/model_card/layer_images/layer_4_attention_self_key.png differ diff --git a/model_card/layer_images/layer_4_attention_self_query.png b/model_card/layer_images/layer_4_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..7095e4dc35512418b322d7256227740d4c7142d8 Binary files /dev/null and b/model_card/layer_images/layer_4_attention_self_query.png differ diff --git a/model_card/layer_images/layer_4_attention_self_value.png b/model_card/layer_images/layer_4_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..95e0fcba3b458436b6e2e993ae86c3fc62504e85 Binary files /dev/null and b/model_card/layer_images/layer_4_attention_self_value.png differ diff --git a/model_card/layer_images/layer_4_intermediate_dense.png b/model_card/layer_images/layer_4_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..60c1cf91adfe74d70390fec77e3f240d87c98342 Binary files /dev/null and b/model_card/layer_images/layer_4_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_4_output_dense.png b/model_card/layer_images/layer_4_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..d36b399a30b18ad1e45a8d7093cc81a6357dd41e Binary files /dev/null and b/model_card/layer_images/layer_4_output_dense.png differ diff --git a/model_card/layer_images/layer_5_attention_output_dense.png b/model_card/layer_images/layer_5_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..f0616d6004ecd5bca0b5ab3e3013ea551b2010df Binary files /dev/null and b/model_card/layer_images/layer_5_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_5_attention_self_key.png b/model_card/layer_images/layer_5_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..bc074ad7c771f7c31b6f0a7dbe05c23c8c560097 Binary files /dev/null and b/model_card/layer_images/layer_5_attention_self_key.png differ diff --git a/model_card/layer_images/layer_5_attention_self_query.png b/model_card/layer_images/layer_5_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..e573960a961616c510f8062e3d49c836529ccfad Binary files /dev/null and b/model_card/layer_images/layer_5_attention_self_query.png differ diff --git a/model_card/layer_images/layer_5_attention_self_value.png b/model_card/layer_images/layer_5_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..137eaeb60241297474f3aa187f24d4df310fc780 Binary files /dev/null and b/model_card/layer_images/layer_5_attention_self_value.png differ diff --git a/model_card/layer_images/layer_5_intermediate_dense.png b/model_card/layer_images/layer_5_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..5333967ae7c6ecdb8d09539da7a3670c2de81560 Binary files /dev/null and b/model_card/layer_images/layer_5_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_5_output_dense.png b/model_card/layer_images/layer_5_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..38b847847fc63a2824f3fdb3a994df6dd838f5d9 Binary files /dev/null and b/model_card/layer_images/layer_5_output_dense.png differ diff --git a/model_card/layer_images/layer_6_attention_output_dense.png b/model_card/layer_images/layer_6_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..dd0dd4244d15c6e2c006d817067040b3365896ef Binary files /dev/null and b/model_card/layer_images/layer_6_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_6_attention_self_key.png b/model_card/layer_images/layer_6_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..823a9659192fe5a322e161a0d631172519807f09 Binary files /dev/null and b/model_card/layer_images/layer_6_attention_self_key.png differ diff --git a/model_card/layer_images/layer_6_attention_self_query.png b/model_card/layer_images/layer_6_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..2a97d21517cdc4db807a2b583f79d22cf30ec2d1 Binary files /dev/null and b/model_card/layer_images/layer_6_attention_self_query.png differ diff --git a/model_card/layer_images/layer_6_attention_self_value.png b/model_card/layer_images/layer_6_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..a08b9cef20e05080e28bbeec89b3db53c84d445d Binary files /dev/null and b/model_card/layer_images/layer_6_attention_self_value.png differ diff --git a/model_card/layer_images/layer_6_intermediate_dense.png b/model_card/layer_images/layer_6_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..41493456475dc6e982726e5311cd9a1bf017f318 Binary files /dev/null and b/model_card/layer_images/layer_6_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_6_output_dense.png b/model_card/layer_images/layer_6_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..79910a0da6bea484f1098aa0d7eb053d8d57d445 Binary files /dev/null and b/model_card/layer_images/layer_6_output_dense.png differ diff --git a/model_card/layer_images/layer_7_attention_output_dense.png b/model_card/layer_images/layer_7_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..ae25f0b12c320d5fbf0fe09a34662cdae99342e3 Binary files /dev/null and b/model_card/layer_images/layer_7_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_7_attention_self_key.png b/model_card/layer_images/layer_7_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..688454d2d1a681941dfd0e26d2126542eaed0abb Binary files /dev/null and b/model_card/layer_images/layer_7_attention_self_key.png differ diff --git a/model_card/layer_images/layer_7_attention_self_query.png b/model_card/layer_images/layer_7_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..6c19bf0cafc5e14d825149bb07241fbec39aad5e Binary files /dev/null and b/model_card/layer_images/layer_7_attention_self_query.png differ diff --git a/model_card/layer_images/layer_7_attention_self_value.png b/model_card/layer_images/layer_7_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..d1b713c8660c05f88acca9fca965fe90ebaf1274 Binary files /dev/null and b/model_card/layer_images/layer_7_attention_self_value.png differ diff --git a/model_card/layer_images/layer_7_intermediate_dense.png b/model_card/layer_images/layer_7_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..7da1afc1a4dbd6ca788adb7ea00e4effe205393e Binary files /dev/null and b/model_card/layer_images/layer_7_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_7_output_dense.png b/model_card/layer_images/layer_7_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..bae7dd1958f0ed7b2f7a09538ecf0f2776539498 Binary files /dev/null and b/model_card/layer_images/layer_7_output_dense.png differ diff --git a/model_card/layer_images/layer_8_attention_output_dense.png b/model_card/layer_images/layer_8_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..973e286b1939dc7bd8bb1069e8db4ccca37bdadb Binary files /dev/null and b/model_card/layer_images/layer_8_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_8_attention_self_key.png b/model_card/layer_images/layer_8_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..598b5cae4d72a1eb7267f36efce11aabd9d63f65 Binary files /dev/null and b/model_card/layer_images/layer_8_attention_self_key.png differ diff --git a/model_card/layer_images/layer_8_attention_self_query.png b/model_card/layer_images/layer_8_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..47d90fb542180bef790611f6c423a67383cdd303 Binary files /dev/null and b/model_card/layer_images/layer_8_attention_self_query.png differ diff --git a/model_card/layer_images/layer_8_attention_self_value.png b/model_card/layer_images/layer_8_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..8ebee1443cfea035449ee3fc56ecaef10350a087 Binary files /dev/null and b/model_card/layer_images/layer_8_attention_self_value.png differ diff --git a/model_card/layer_images/layer_8_intermediate_dense.png b/model_card/layer_images/layer_8_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..24ba7e7d1ddf6ab42ec6a0148ad173dd6a4ec9b4 Binary files /dev/null and b/model_card/layer_images/layer_8_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_8_output_dense.png b/model_card/layer_images/layer_8_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..644b89b1bac2fff3dee9cc690d1cf3ec47ee3a68 Binary files /dev/null and b/model_card/layer_images/layer_8_output_dense.png differ diff --git a/model_card/layer_images/layer_9_attention_output_dense.png b/model_card/layer_images/layer_9_attention_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..14b2fa41d7dcb4f1a36d89fbfd4adaaf8ec1d0ad Binary files /dev/null and b/model_card/layer_images/layer_9_attention_output_dense.png differ diff --git a/model_card/layer_images/layer_9_attention_self_key.png b/model_card/layer_images/layer_9_attention_self_key.png new file mode 100644 index 0000000000000000000000000000000000000000..ed4e09634fd5b806dd9f3422ec1572c56a4a92c7 Binary files /dev/null and b/model_card/layer_images/layer_9_attention_self_key.png differ diff --git a/model_card/layer_images/layer_9_attention_self_query.png b/model_card/layer_images/layer_9_attention_self_query.png new file mode 100644 index 0000000000000000000000000000000000000000..db88700555486ae65efb8e3492e861a14d67cd20 Binary files /dev/null and b/model_card/layer_images/layer_9_attention_self_query.png differ diff --git a/model_card/layer_images/layer_9_attention_self_value.png b/model_card/layer_images/layer_9_attention_self_value.png new file mode 100644 index 0000000000000000000000000000000000000000..5d47e08daaa8ba51345dfbe08f0047b2a844c680 Binary files /dev/null and b/model_card/layer_images/layer_9_attention_self_value.png differ diff --git a/model_card/layer_images/layer_9_intermediate_dense.png b/model_card/layer_images/layer_9_intermediate_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..0e4e5aa024c7d66cf8050cec2c82b24e01b1d918 Binary files /dev/null and b/model_card/layer_images/layer_9_intermediate_dense.png differ diff --git a/model_card/layer_images/layer_9_output_dense.png b/model_card/layer_images/layer_9_output_dense.png new file mode 100644 index 0000000000000000000000000000000000000000..1c4d4015bd7acbe7a8ba7fddbbb60c005d14c300 Binary files /dev/null and b/model_card/layer_images/layer_9_output_dense.png differ diff --git a/model_card/pruning.svg b/model_card/pruning.svg new file mode 100644 index 0000000000000000000000000000000000000000..16f42429f313084d7833b66cb0feff259f424a4a --- /dev/null +++ b/model_card/pruning.svg @@ -0,0 +1 @@ +33363455353499969877979801234567891011024681012prunedactivePruned Transformer HeadsLayer indexHeads count \ No newline at end of file diff --git a/pytorch_model.bin b/pytorch_model.bin index 825e8e5b8061c0083533e620c5f064f576a0a7ac..8f1a9e16354ae3c61647523c9c9dedb978b5d8d7 100644 --- a/pytorch_model.bin +++ b/pytorch_model.bin @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:bff0440beecbc2d32a8b2a7a85b9a4fbe76bedfa9768334b5c613ba8831310ea +oid sha256:8ed1e49619c312eb77c5841ef896fe234c81217368becf7dc0a3fdaeb0d1be17 size 359300023