@elgap/edukaai 0.1.0

package/.output/server/chunks/build/glossary-CxSGCJoH.mjs

@@ -0,0 +1,397 @@
+import { _ as __nuxt_component_0 } from './nuxt-link-Ceyd90PQ.mjs';
+import { computed, mergeProps, withCtx, createTextVNode, unref, toDisplayString, useSSRContext } from 'vue';
+import { ssrRenderAttrs, ssrRenderComponent, ssrInterpolate, ssrRenderList } from 'vue/server-renderer';
+import { b as useRoute } from './server.mjs';
+import '../nitro/nitro.mjs';
+import 'node:http';
+import 'node:https';
+import 'node:events';
+import 'node:buffer';
+import 'node:fs';
+import 'node:path';
+import 'node:crypto';
+import 'node:url';
+import '@iconify/utils';
+import 'consola';
+import 'pinia';
+import 'vue-router';
+import 'tailwindcss/colors';
+import '@iconify/vue';
+import '../routes/renderer.mjs';
+import 'vue-bundle-renderer/runtime';
+import 'unhead/server';
+import 'devalue';
+import 'unhead/utils';
+
+const _sfc_main = {
+  __name: "glossary",
+  __ssrInlineRender: true,
+  setup(__props) {
+    const route = useRoute();
+    const glossaryTerms = {
+      "base-model": {
+        emoji: "🏗️",
+        word: "Base Model",
+        shortDefinition: "The pre-trained LLM you start with before fine-tuning (e.g., GPT-4, Llama 2, Mistral).",
+        detailedExplanation: `
+      <p class="mb-4">A <strong>base model</strong> (also called a "foundation model" or "pre-trained model") is the starting point for fine-tuning. It's an LLM that has already been trained on massive amounts of text data (often trillions of tokens from books, websites, and documents).</p>
+      
+      <h3 class="text-lg font-semibold mt-6 mb-3">Key Characteristics:</h3>
+      <ul class="list-disc list-inside space-y-2 mb-4">
+        <li>Knows general language patterns, grammar, and facts</li>
+        <li>Can answer general questions and write coherent text</li>
+        <li>Hasn't been specialized for any specific domain</li>
+        <li>Expensive and time-consuming to train from scratch (weeks, millions of dollars)</li>
+        <li>Usually released openly (open source) or via API (closed source)</li>
+      </ul>
+
+      <h3 class="text-lg font-semibold mt-6 mb-3">Popular Base Models:</h3>
+      <div class="grid grid-cols-1 md:grid-cols-2 gap-3 mb-4">
+        <div class="p-3 bg-gray-50 rounded-lg">
+          <strong class="text-blue-700">Llama 2 (Meta)</strong>
+          <p class="text-sm text-gray-600 mt-1">Open source, 7B-70B parameters, great for fine-tuning</p>
+        </div>
+        <div class="p-3 bg-gray-50 rounded-lg">
+          <strong class="text-blue-700">Mistral</strong>
+          <p class="text-sm text-gray-600 mt-1">Open source, very efficient, good performance</p>
+        </div>
+        <div class="p-3 bg-gray-50 rounded-lg">
+          <strong class="text-blue-700">GPT-3.5 (OpenAI)</strong>
+          <p class="text-sm text-gray-600 mt-1">Closed source, available via fine-tuning API</p>
+        </div>
+        <div class="p-3 bg-gray-50 rounded-lg">
+          <strong class="text-blue-700">Falcon</strong>
+          <p class="text-sm text-gray-600 mt-1">Open source, trained on high-quality data</p>
+        </div>
+      </div>
+
+      <h3 class="text-lg font-semibold mt-6 mb-3">Why Base Models Matter:</h3>
+      <p class="mb-4">When you fine-tune, you're leveraging all the knowledge the base model learned during its expensive pre-training. You're teaching it your specific domain while preserving its general capabilities. A good base model means less work for you and better results.</p>
+
+      <div class="bg-yellow-50 p-4 rounded-lg mt-6">
+        <strong class="text-yellow-800">💡 Pro Tip:</strong>
+        <p class="text-sm text-yellow-700 mt-1">For beginners, start with Llama 2 (7B or 13B parameters). They're free, well-documented, and perfect for learning. As you get more experienced, you can experiment with larger models.</p>
+      </div>
+    `,
+        example: "You download Llama 2 7B (an open-source base model) and fine-tune it on 1000 examples of customer support conversations. Now it specializes in helping customers while retaining its general language abilities.",
+        seeAlso: ["fine-tuning", "parameters", "training"]
+      },
+      "dataset": {
+        emoji: "📊",
+        word: "Dataset",
+        shortDefinition: "A collection of training examples used to teach the AI specific tasks.",
+        detailedExplanation: `
+      <p class="mb-4">A <strong>dataset</strong> is the fuel that powers fine-tuning. It's a structured collection of examples showing the model what you want it to learn. Each example contains an input (instruction) and the desired output (response).</p>
+      
+      <h3 class="text-lg font-semibold mt-6 mb-3">Dataset Structure:</h3>
+      <div class="bg-gray-50 p-4 rounded-lg mb-4">
+        <p class="font-mono text-sm mb-2">Example format:</p>
+        <div class="space-y-2 text-sm">
+          <div class="p-2 bg-white rounded border-l-4 border-blue-500">
+            <strong>Instruction:</strong> "How do I reverse a list in Python?"
+          </div>
+          <div class="p-2 bg-white rounded border-l-4 border-green-500">
+            <strong>Output:</strong> "You can use list slicing: my_list[::-1] or the reversed() function."
+          </div>
+        </div>
+      </div>
+
+      <h3 class="text-lg font-semibold mt-6 mb-3">What Makes a Good Dataset:</h3>
+      <ul class="list-disc list-inside space-y-2 mb-4">
+        <li><strong>Quality over quantity</strong> — 100 excellent examples beat 1000 mediocre ones</li>
+        <li><strong>Diversity</strong> — Cover different question types, formats, and edge cases</li>
+        <li><strong>Consistency</strong> — Similar tone and style across examples</li>
+        <li><strong>Accuracy</strong> — Factually correct information</li>
+        <li><strong>Completeness</strong> — Examples should fully answer the question</li>
+      </ul>
+
+      <h3 class="text-lg font-semibold mt-6 mb-3">Dataset Size Guidelines:</h3>
+      <table class="w-full text-sm mb-4">
+        <thead>
+          <tr class="border-b">
+            <th class="text-left py-2">Size</th>
+            <th class="text-left py-2">Use Case</th>
+            <th class="text-left py-2">Expected Results</th>
+          </tr>
+        </thead>
+        <tbody>
+          <tr class="border-b">
+            <td class="py-2">100 examples</td>
+            <td>Testing/prototyping</td>
+            <td>Minimal viable dataset</td>
+          </tr>
+          <tr class="border-b">
+            <td class="py-2">500 examples</td>
+            <td>Simple tasks</td>
+            <td>Noticeable improvement</td>
+          </tr>
+          <tr class="border-b">
+            <td class="py-2">1000 examples</td>
+            <td>Most use cases</td>
+            <td>Significant improvement</td>
+          </tr>
+          <tr>
+            <td class="py-2">5000+ examples</td>
+            <td>Professional use</td>
+            <td>High-quality specialized model</td>
+          </tr>
+        </tbody>
+      </table>
+
+      <div class="bg-blue-50 p-4 rounded-lg mt-6">
+        <strong class="text-blue-800">🎯 The 1000 Example Milestone:</strong>
+        <p class="text-sm text-blue-700 mt-1">This is the sweet spot for beginners. It's enough data for the model to learn patterns, but not so much that training takes forever. At 20-30 examples per day, you'll reach this in 2-3 months.</p>
+      </div>
+    `,
+        example: "You collect 1000 examples of customer service conversations. Each example shows a customer question and the ideal support agent response. This becomes your training dataset.",
+        seeAlso: ["training-example", "fine-tuning", "training"]
+      },
+      "epoch": {
+        emoji: "🔄",
+        word: "Epoch",
+        shortDefinition: "One complete pass through the entire dataset during training.",
+        detailedExplanation: `
+      <p class="mb-4">An <strong>epoch</strong> is when the model has seen every example in your dataset exactly once. Think of it like reading a textbook cover to cover — one epoch is one complete read-through.</p>
+
+      <h3 class="text-lg font-semibold mt-6 mb-3">Why Multiple Epochs?</h3>
+      <p class="mb-4">Just like you don't fully understand a textbook after one reading, the model doesn't fully learn from one epoch. Multiple passes help it:</p>
+      <ul class="list-disc list-inside space-y-2 mb-4">
+        <li>Reinforce patterns it learned</li>
+        <li>Pick up on subtle details missed the first time</li>
+        <li>Build stronger associations between concepts</li>
+        <li>Gradually improve its predictions</li>
+      </ul>
+
+      <h3 class="text-lg font-semibold mt-6 mb-3">Typical Epoch Counts:</h3>
+      <div class="space-y-2 mb-4">
+        <div class="flex justify-between items-center p-2 bg-gray-50 rounded">
+          <span>Pre-training (from scratch)</span>
+          <span class="font-mono">1-3 epochs</span>
+        </div>
+        <div class="flex justify-between items-center p-2 bg-gray-50 rounded">
+          <span>Fine-tuning (medium dataset)</span>
+          <span class="font-mono">3-10 epochs</span>
+        </div>
+        <div class="flex justify-between items-center p-2 bg-gray-50 rounded">
+          <span>Fine-tuning (small dataset)</span>
+          <span class="font-mono">10-30 epochs</span>
+        </div>
+      </div>
+
+      <h3 class="text-lg font-semibold mt-6 mb-3">The Trade-off:</h3>
+      <div class="grid grid-cols-1 md:grid-cols-2 gap-4 mb-4">
+        <div class="p-3 bg-green-50 rounded-lg border border-green-200">
+          <strong class="text-green-700">More Epochs ✅</strong>
+          <ul class="text-sm text-gray-700 mt-2 space-y-1">
+            <li>Better learning</li>
+            <li>Stronger patterns</li>
+            <li>Higher accuracy</li>
+          </ul>
+        </div>
+        <div class="p-3 bg-red-50 rounded-lg border border-red-200">
+          <strong class="text-red-700">Too Many Epochs ❌</strong>
+          <ul class="text-sm text-gray-700 mt-2 space-y-1">
+            <li>Overfitting (memorization)</li>
+            <li>Longer training time</li>
+            <li>Higher cost</li>
+          </ul>
+        </div>
+      </div>
+
+      <div class="bg-yellow-50 p-4 rounded-lg mt-6">
+        <strong class="text-yellow-800">⚠️ Overfitting Warning:</strong>
+        <p class="text-sm text-yellow-700 mt-1">If you train for too many epochs on a small dataset, the model might memorize the examples instead of learning the underlying patterns. It will perform perfectly on training data but fail on new inputs. This is why validation sets are important!</p>
+      </div>
+    `,
+        example: "You have 1000 training examples. After 1 epoch, the model has seen each example once. After 3 epochs, it has seen them all three times. Most fine-tuning jobs use 3-10 epochs.",
+        seeAlso: ["training", "fine-tuning", "overfitting"]
+      },
+      "fine-tuning": {
+        emoji: "🎓",
+        word: "Fine-Tuning",
+        shortDefinition: "Training a pre-trained model on a specific dataset to improve performance on particular tasks.",
+        detailedExplanation: `
+      <p class="mb-4"><strong>Fine-tuning</strong> is the process of taking a pre-trained LLM (which knows general language and facts) and teaching it to be better at your specific task or domain. It's like giving a college graduate specialized job training.</p>
+
+      <h3 class="text-lg font-semibold mt-6 mb-3">The Fine-Tuning Process:</h3>
+      <div class="space-y-4 mb-6">
+        <div class="flex items-start gap-3">
+          <div class="w-8 h-8 bg-blue-500 rounded-full flex items-center justify-center text-white font-bold text-sm flex-shrink-0">1</div>
+          <div>
+            <strong>Start with Base Model</strong>
+            <p class="text-sm text-gray-600">You begin with a pre-trained model like Llama 2 or GPT-3.5</p>
+          </div>
+        </div>
+        <div class="flex items-start gap-3">
+          <div class="w-8 h-8 bg-blue-500 rounded-full flex items-center justify-center text-white font-bold text-sm flex-shrink-0">2</div>
+          <div>
+            <strong>Prepare Dataset</strong>
+            <p class="text-sm text-gray-600">Collect 100-10000 examples of your specific task</p>
+          </div>
+        </div>
+        <div class="flex items-start gap-3">
+          <div class="w-8 h-8 bg-blue-500 rounded-full flex items-center justify-center text-white font-bold text-sm flex-shrink-0">3</div>
+          <div>
+            <strong>Continue Training</strong>
+            <p class="text-sm text-gray-600">Train the model on your dataset for several epochs</p>
+          </div>
+        </div>
+        <div class="flex items-start gap-3">
+          <div class="w-8 h-8 bg-blue-500 rounded-full flex items-center justify-center text-white font-bold text-sm flex-shrink-0">4</div>
+          <div>
+            <strong>Deploy & Use</strong>
+            <p class="text-sm text-gray-600">Your model now specializes in your domain</p>
+          </div>
+        </div>
+      </div>
+
+      <h3 class="text-lg font-semibold mt-6 mb-3">Benefits of Fine-Tuning:</h3>
+      <div class="grid grid-cols-1 md:grid-cols-2 gap-4 mb-4">
+        <div class="p-3 bg-green-50 rounded-lg">
+          <strong class="text-green-700">✅ Better Performance</strong>
+          <p class="text-sm text-gray-600 mt-1">Expert-level responses in your specific domain</p>
+        </div>
+        <div class="p-3 bg-blue-50 rounded-lg">
+          <strong class="text-blue-700">✅ Lower Costs</strong>
+          <p class="text-sm text-gray-600 mt-1">Smaller fine-tuned models can beat larger general models</p>
+        </div>
+        <div class="p-3 bg-purple-50 rounded-lg">
+          <strong class="text-purple-700">✅ Consistent Style</strong>
+          <p class="text-sm text-gray-600 mt-1">Train your brand voice and tone</p>
+        </div>
+        <div class="p-3 bg-orange-50 rounded-lg">
+          <strong class="text-orange-700">✅ Privacy</strong>
+          <p class="text-sm text-gray-600 mt-1">Run on your own infrastructure, no API calls</p>
+        </div>
+      </div>
+
+      <h3 class="text-lg font-semibold mt-6 mb-3">Fine-Tuning vs. Pre-training:</h3>
+      <table class="w-full text-sm mb-4">
+        <thead>
+          <tr class="border-b">
+            <th class="text-left py-2">Aspect</th>
+            <th class="text-left py-2">Pre-training</th>
+            <th class="text-left py-2">Fine-tuning</th>
+          </tr>
+        </thead>
+        <tbody>
+          <tr class="border-b">
+            <td class="py-2">Data needed</td>
+            <td class="py-2">Trillions of tokens</td>
+            <td class="py-2">Thousands to millions</td>
+          </tr>
+          <tr class="border-b">
+            <td class="py-2">Time required</td>
+            <td class="py-2">Weeks/months</td>
+            <td class="py-2">Hours to days</td>
+          </tr>
+          <tr class="border-b">
+            <td class="py-2">Cost</td>
+            <td class="py-2">Millions of dollars</td>
+            <td class="py-2">$10-$1000</td>
+          </tr>
+          <tr>
+            <td class="py-2">Result</td>
+            <td class="py-2">General knowledge</td>
+            <td class="py-2">Domain expertise</td>
+          </tr>
+        </tbody>
+      </table>
+
+      <div class="bg-green-50 p-4 rounded-lg mt-6">
+        <strong class="text-green-800">🎯 Perfect for Beginners:</strong>
+        <p class="text-sm text-green-700 mt-1">Fine-tuning is why edukaAI exists. Instead of training from scratch (impossible for individuals), you leverage pre-trained models and just teach them your specialty. It's affordable, fast, and achievable!</p>
+      </div>
+    `,
+        example: "You take Llama 2 (general model) and fine-tune it on 1000 customer support conversations. Now it understands your products, tone, and common issues, while retaining its general language abilities.",
+        seeAlso: ["base-model", "training", "dataset", "training-example"]
+      }
+    };
+    const termId = route.query.term;
+    const term = computed(() => glossaryTerms[termId]);
+    const getTermTitle = (id) => {
+      return glossaryTerms[id]?.word || id;
+    };
+    return (_ctx, _push, _parent, _attrs) => {
+      const _component_NuxtLink = __nuxt_component_0;
+      _push(`<div${ssrRenderAttrs(mergeProps({ class: "max-w-4xl mx-auto pb-20" }, _attrs))}><div class="text-sm text-gray-500 mb-6">`);
+      _push(ssrRenderComponent(_component_NuxtLink, {
+        to: "/help",
+        class: "hover:text-blue-600"
+      }, {
+        default: withCtx((_, _push2, _parent2, _scopeId) => {
+          if (_push2) {
+            _push2(`← Back to Guide`);
+          } else {
+            return [
+              createTextVNode("← Back to Guide")
+            ];
+          }
+        }),
+        _: 1
+      }, _parent));
+      _push(`</div>`);
+      if (unref(term)) {
+        _push(`<div class="mb-12"><div class="flex items-center gap-3 mb-4"><span class="text-4xl">${ssrInterpolate(unref(term).emoji)}</span><h1 class="text-4xl font-bold">${ssrInterpolate(unref(term).word)}</h1></div><div class="card mb-6"><h2 class="text-lg font-semibold mb-3 text-blue-700">Quick Definition</h2><p class="text-lg text-gray-700">${ssrInterpolate(unref(term).shortDefinition)}</p></div><div class="card"><h2 class="text-lg font-semibold mb-4 text-blue-700">Detailed Explanation</h2><div class="prose max-w-none text-gray-700">${unref(term).detailedExplanation ?? ""}</div>`);
+        if (unref(term).example) {
+          _push(`<div class="mt-6 p-4 bg-gray-50 rounded-lg"><h3 class="font-medium mb-2">💡 Example</h3><p class="text-sm text-gray-700">${ssrInterpolate(unref(term).example)}</p></div>`);
+        } else {
+          _push(`<!---->`);
+        }
+        if (unref(term).seeAlso && unref(term).seeAlso.length) {
+          _push(`<div class="mt-6 pt-6 border-t"><h3 class="font-medium mb-2">🔗 See Also</h3><div class="flex flex-wrap gap-2"><!--[-->`);
+          ssrRenderList(unref(term).seeAlso, (related) => {
+            _push(ssrRenderComponent(_component_NuxtLink, {
+              key: related,
+              to: `/help/glossary?term=${related}`,
+              class: "px-3 py-1 bg-blue-100 text-blue-700 rounded-full text-sm hover:bg-blue-200 transition-colors"
+            }, {
+              default: withCtx((_, _push2, _parent2, _scopeId) => {
+                if (_push2) {
+                  _push2(`${ssrInterpolate(getTermTitle(related))}`);
+                } else {
+                  return [
+                    createTextVNode(toDisplayString(getTermTitle(related)), 1)
+                  ];
+                }
+              }),
+              _: 2
+            }, _parent));
+          });
+          _push(`<!--]--></div></div>`);
+        } else {
+          _push(`<!---->`);
+        }
+        _push(`</div></div>`);
+      } else {
+        _push(`<div class="text-center py-12"><p class="text-gray-500">Select a term from the glossary</p>`);
+        _push(ssrRenderComponent(_component_NuxtLink, {
+          to: "/help",
+          class: "btn-primary mt-4 inline-block"
+        }, {
+          default: withCtx((_, _push2, _parent2, _scopeId) => {
+            if (_push2) {
+              _push2(`Back to Guide`);
+            } else {
+              return [
+                createTextVNode("Back to Guide")
+              ];
+            }
+          }),
+          _: 1
+        }, _parent));
+        _push(`</div>`);
+      }
+      _push(`</div>`);
+    };
+  }
+};
+const _sfc_setup = _sfc_main.setup;
+_sfc_main.setup = (props, ctx) => {
+  const ssrContext = useSSRContext();
+  (ssrContext.modules || (ssrContext.modules = /* @__PURE__ */ new Set())).add("pages/help/glossary.vue");
+  return _sfc_setup ? _sfc_setup(props, ctx) : void 0;
+};
+
+export { _sfc_main as default };
+//# sourceMappingURL=glossary-CxSGCJoH.mjs.map

package/.output/server/chunks/build/glossary-CxSGCJoH.mjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"glossary-CxSGCJoH.mjs","sources":["../../../../node_modules/.cache/nuxt/.nuxt/dist/server/_nuxt/glossary-CxSGCJoH.js"],"names":[],"mappings":"","x_google_ignoreList":[0]}

package/.output/server/chunks/build/huggingface-TSmyUzZU.mjs

@@ -0,0 +1,128 @@
+import { _ as __nuxt_component_0 } from './nuxt-link-Ceyd90PQ.mjs';
+import { mergeProps, withCtx, createTextVNode, createVNode, useSSRContext } from 'vue';
+import { ssrRenderAttrs, ssrRenderComponent } from 'vue/server-renderer';
+import '../nitro/nitro.mjs';
+import 'node:http';
+import 'node:https';
+import 'node:events';
+import 'node:buffer';
+import 'node:fs';
+import 'node:path';
+import 'node:crypto';
+import 'node:url';
+import '@iconify/utils';
+import 'consola';
+import './server.mjs';
+import 'pinia';
+import 'vue-router';
+import 'tailwindcss/colors';
+import '@iconify/vue';
+import '../routes/renderer.mjs';
+import 'vue-bundle-renderer/runtime';
+import 'unhead/server';
+import 'devalue';
+import 'unhead/utils';
+
+const _sfc_main = {
+  __name: "huggingface",
+  __ssrInlineRender: true,
+  setup(__props) {
+    return (_ctx, _push, _parent, _attrs) => {
+      const _component_NuxtLink = __nuxt_component_0;
+      _push(`<div${ssrRenderAttrs(mergeProps({ class: "max-w-4xl mx-auto" }, _attrs))}><div class="mb-8"><div class="flex items-center gap-3 mb-4">`);
+      _push(ssrRenderComponent(_component_NuxtLink, {
+        to: "/export",
+        class: "text-blue-600 hover:text-blue-800"
+      }, {
+        default: withCtx((_, _push2, _parent2, _scopeId) => {
+          if (_push2) {
+            _push2(` ← Back to Export `);
+          } else {
+            return [
+              createTextVNode(" ← Back to Export ")
+            ];
+          }
+        }),
+        _: 1
+      }, _parent));
+      _push(`</div><h1 class="text-3xl font-bold mb-4">🤗 What is Hugging Face?</h1><p class="text-xl text-gray-600"> The &quot;GitHub of AI&quot; - where the machine learning community shares models, datasets, and tools </p></div><div class="card mb-6"><h2 class="text-xl font-semibold mb-4">In Simple Terms</h2><div class="space-y-4 text-gray-700"><p><strong>Hugging Face</strong> is like GitHub, but specifically for artificial intelligence and machine learning. It&#39;s the largest community hub where researchers, developers, and companies share: </p><div class="grid grid-cols-1 md:grid-cols-3 gap-4 my-6"><div class="p-4 bg-yellow-50 rounded-lg border border-yellow-200"><div class="text-3xl mb-2">🤖</div><h3 class="font-semibold mb-1">Pre-trained Models</h3><p class="text-sm text-gray-600"> Thousands of ready-to-use AI models (GPT, LLaMA, BERT, etc.) </p></div><div class="p-4 bg-green-50 rounded-lg border border-green-200"><div class="text-3xl mb-2">📊</div><h3 class="font-semibold mb-1">Datasets</h3><p class="text-sm text-gray-600"> Curated datasets for training and benchmarking </p></div><div class="p-4 bg-blue-50 rounded-lg border border-blue-200"><div class="text-3xl mb-2">🛠️</div><h3 class="font-semibold mb-1">Tools &amp; Libraries</h3><p class="text-sm text-gray-600"> Transformers, Datasets, and other ML tools </p></div></div><p> Think of it this way: if you want to build a website, you might use GitHub to find code. If you want to build AI, you use Hugging Face to find models and data. </p></div></div><div class="card mb-6"><h2 class="text-xl font-semibold mb-4">Why Use Hugging Face for Fine-tuning?</h2><div class="space-y-4"><div class="flex gap-4 items-start"><div class="w-8 h-8 bg-blue-100 rounded-full flex items-center justify-center flex-shrink-0"><span class="text-blue-600 font-bold">1</span></div><div><h3 class="font-semibold mb-1">Access to Base Models</h3><p class="text-gray-600"> Download foundation models like LLaMA, Mistral, or GPT-2 to start your fine-tuning </p></div></div><div class="flex gap-4 items-start"><div class="w-8 h-8 bg-blue-100 rounded-full flex items-center justify-center flex-shrink-0"><span class="text-blue-600 font-bold">2</span></div><div><h3 class="font-semibold mb-1">Share Your Trained Models</h3><p class="text-gray-600"> After fine-tuning, upload your model so others can use it (or keep it private) </p></div></div><div class="flex gap-4 items-start"><div class="w-8 h-8 bg-blue-100 rounded-full flex items-center justify-center flex-shrink-0"><span class="text-blue-600 font-bold">3</span></div><div><h3 class="font-semibold mb-1">Inference API</h3><p class="text-gray-600"> Test your models online without setting up infrastructure </p></div></div><div class="flex gap-4 items-start"><div class="w-8 h-8 bg-blue-100 rounded-full flex items-center justify-center flex-shrink-0"><span class="text-blue-600 font-bold">4</span></div><div><h3 class="font-semibold mb-1">Community &amp; Documentation</h3><p class="text-gray-600"> Extensive docs, tutorials, and a helpful community for beginners </p></div></div></div></div><div class="card mb-6"><h2 class="text-xl font-semibold mb-4">Key Concepts to Know</h2><div class="space-y-4"><div class="p-4 bg-gray-50 rounded-lg"><h3 class="font-semibold mb-2">🏠 Model Hub</h3><p class="text-gray-600 text-sm"> Browse and download thousands of pre-trained models. Each model page shows: model size, license, community ratings, and example usage code. </p></div><div class="p-4 bg-gray-50 rounded-lg"><h3 class="font-semibold mb-2">📁 Datasets</h3><p class="text-gray-600 text-sm"> Search for training datasets by task (text generation, classification, etc.). You can also upload your edukaAI dataset here. </p></div><div class="p-4 bg-gray-50 rounded-lg"><h3 class="font-semibold mb-2">🎪 Spaces</h3><p class="text-gray-600 text-sm"> Interactive demos of AI models. Great for testing models before downloading. </p></div><div class="p-4 bg-gray-50 rounded-lg"><h3 class="font-semibold mb-2">🐛 Transformers Library</h3><p class="text-gray-600 text-sm"> Python library for downloading and using models. The main tool for fine-tuning. </p></div></div></div><div class="card mb-6"><h2 class="text-xl font-semibold mb-4">Getting Started with Fine-tuning</h2><div class="space-y-4"><p class="text-gray-700"> Here&#39;s the typical workflow when using Hugging Face for fine-tuning: </p><div class="flex items-center gap-4 py-4"><div class="flex-1 text-center"><div class="w-12 h-12 bg-blue-100 rounded-full flex items-center justify-center mx-auto mb-2"><span class="text-xl">📥</span></div><p class="text-sm font-medium">1. Download Base Model</p><p class="text-xs text-gray-500">From Model Hub</p></div><div class="text-2xl text-gray-300">→</div><div class="flex-1 text-center"><div class="w-12 h-12 bg-green-100 rounded-full flex items-center justify-center mx-auto mb-2"><span class="text-xl">📊</span></div><p class="text-sm font-medium">2. Prepare Dataset</p><p class="text-xs text-gray-500">Use edukaAI!</p></div><div class="text-2xl text-gray-300">→</div><div class="flex-1 text-center"><div class="w-12 h-12 bg-purple-100 rounded-full flex items-center justify-center mx-auto mb-2"><span class="text-xl">⚙️</span></div><p class="text-sm font-medium">3. Fine-tune</p><p class="text-xs text-gray-500">Train the model</p></div><div class="text-2xl text-gray-300">→</div><div class="flex-1 text-center"><div class="w-12 h-12 bg-yellow-100 rounded-full flex items-center justify-center mx-auto mb-2"><span class="text-xl">📤</span></div><p class="text-sm font-medium">4. Upload Result</p><p class="text-xs text-gray-500">Share your model</p></div></div></div></div><div class="card mb-6 bg-gradient-to-r from-yellow-50 to-orange-50 border-yellow-200"><h2 class="text-xl font-semibold mb-4">⚠️ What to Expect</h2><div class="space-y-3 text-gray-700"><p><strong>Hugging Face is powerful but can be overwhelming for beginners.</strong> Here&#39;s what you should know: </p><ul class="space-y-2 ml-4"><li class="flex items-start gap-2"><span class="text-yellow-600">•</span><span>Requires Python knowledge and command line comfort</span></li><li class="flex items-start gap-2"><span class="text-yellow-600">•</span><span>Fine-tuning requires GPU (either local or cloud)</span></li><li class="flex items-start gap-2"><span class="text-yellow-600">•</span><span>Documentation is comprehensive but technical</span></li><li class="flex items-start gap-2"><span class="text-yellow-600">•</span><span>Many tutorials assume prior ML knowledge</span></li></ul><p class="mt-4"><strong>💡 Tip:</strong> If you&#39;re completely new, consider using `);
+      _push(ssrRenderComponent(_component_NuxtLink, {
+        to: "/help/axolotl",
+        class: "text-blue-600 hover:underline font-semibold"
+      }, {
+        default: withCtx((_, _push2, _parent2, _scopeId) => {
+          if (_push2) {
+            _push2(`Axolotl`);
+          } else {
+            return [
+              createTextVNode("Axolotl")
+            ];
+          }
+        }),
+        _: 1
+      }, _parent));
+      _push(` first - it simplifies the Hugging Face workflow significantly. </p></div></div><div class="card mb-6"><h2 class="text-xl font-semibold mb-4">Helpful Resources</h2><div class="space-y-3"><a href="https://huggingface.co/docs/transformers/training" target="_blank" class="flex items-center gap-3 p-3 bg-blue-50 rounded-lg hover:bg-blue-100 transition-colors"><span class="text-2xl">📚</span><div class="flex-1"><h3 class="font-semibold">Official Training Guide</h3><p class="text-sm text-gray-600">Hugging Face&#39;s complete fine-tuning tutorial</p></div><span class="text-blue-600">→</span></a><a href="https://huggingface.co/models" target="_blank" class="flex items-center gap-3 p-3 bg-yellow-50 rounded-lg hover:bg-yellow-100 transition-colors"><span class="text-2xl">🤖</span><div class="flex-1"><h3 class="font-semibold">Model Hub</h3><p class="text-sm text-gray-600">Browse thousands of pre-trained models</p></div><span class="text-yellow-600">→</span></a><a href="https://huggingface.co/datasets" target="_blank" class="flex items-center gap-3 p-3 bg-green-50 rounded-lg hover:bg-green-100 transition-colors"><span class="text-2xl">📊</span><div class="flex-1"><h3 class="font-semibold">Datasets</h3><p class="text-sm text-gray-600">Find datasets for your training tasks</p></div><span class="text-green-600">→</span></a>`);
+      _push(ssrRenderComponent(_component_NuxtLink, {
+        to: "/help/llm-training",
+        class: "flex items-center gap-3 p-3 bg-purple-50 rounded-lg hover:bg-purple-100 transition-colors"
+      }, {
+        default: withCtx((_, _push2, _parent2, _scopeId) => {
+          if (_push2) {
+            _push2(`<span class="text-2xl"${_scopeId}>🎓</span><div class="flex-1"${_scopeId}><h3 class="font-semibold"${_scopeId}>edukaAI: How LLM Training Works</h3><p class="text-sm text-gray-600"${_scopeId}>Understand the fundamentals before you start</p></div><span class="text-purple-600"${_scopeId}>→</span>`);
+          } else {
+            return [
+              createVNode("span", { class: "text-2xl" }, "🎓"),
+              createVNode("div", { class: "flex-1" }, [
+                createVNode("h3", { class: "font-semibold" }, "edukaAI: How LLM Training Works"),
+                createVNode("p", { class: "text-sm text-gray-600" }, "Understand the fundamentals before you start")
+              ]),
+              createVNode("span", { class: "text-purple-600" }, "→")
+            ];
+          }
+        }),
+        _: 1
+      }, _parent));
+      _push(`</div></div><div class="card bg-gradient-to-r from-blue-600 to-purple-600 text-white"><h2 class="text-xl font-semibold mb-4">Ready to Start?</h2><p class="mb-6 opacity-90"> You now understand what Hugging Face is and how it fits into the fine-tuning workflow. Remember, you can always come back to edukaAI to create more training data! </p><div class="flex flex-col sm:flex-row gap-3"><a href="https://huggingface.co" target="_blank" class="flex-1 bg-white text-blue-600 font-semibold py-3 px-6 rounded-lg text-center hover:bg-gray-100 transition-colors"> Go to HuggingFace.co → </a>`);
+      _push(ssrRenderComponent(_component_NuxtLink, {
+        to: "/help/axolotl",
+        class: "flex-1 bg-blue-700 text-white font-semibold py-3 px-6 rounded-lg text-center hover:bg-blue-800 transition-colors"
+      }, {
+        default: withCtx((_, _push2, _parent2, _scopeId) => {
+          if (_push2) {
+            _push2(` Learn About Axolotl → `);
+          } else {
+            return [
+              createTextVNode(" Learn About Axolotl → ")
+            ];
+          }
+        }),
+        _: 1
+      }, _parent));
+      _push(ssrRenderComponent(_component_NuxtLink, {
+        to: "/export",
+        class: "flex-1 bg-transparent border-2 border-white text-white font-semibold py-3 px-6 rounded-lg text-center hover:bg-white/10 transition-colors"
+      }, {
+        default: withCtx((_, _push2, _parent2, _scopeId) => {
+          if (_push2) {
+            _push2(` ← Back to Export `);
+          } else {
+            return [
+              createTextVNode(" ← Back to Export ")
+            ];
+          }
+        }),
+        _: 1
+      }, _parent));
+      _push(`</div></div></div>`);
+    };
+  }
+};
+const _sfc_setup = _sfc_main.setup;
+_sfc_main.setup = (props, ctx) => {
+  const ssrContext = useSSRContext();
+  (ssrContext.modules || (ssrContext.modules = /* @__PURE__ */ new Set())).add("pages/help/huggingface.vue");
+  return _sfc_setup ? _sfc_setup(props, ctx) : void 0;
+};
+
+export { _sfc_main as default };
+//# sourceMappingURL=huggingface-TSmyUzZU.mjs.map

package/.output/server/chunks/build/huggingface-TSmyUzZU.mjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"huggingface-TSmyUzZU.mjs","sources":["../../../../node_modules/.cache/nuxt/.nuxt/dist/server/_nuxt/huggingface-TSmyUzZU.js"],"names":[],"mappings":"","x_google_ignoreList":[0]}