tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 3 Packages Projects Releases Wiki Activity

Built site for gh-pages

Browse Source
This commit is contained in:
Quarto GHA Workflow Runner
2025-06-14 18:56:28 +00:00
parent 84db47f3c0
commit 5b66b8e86c
6 changed files with 393 additions and 339 deletions
Download Patch File Download Diff File Expand all files Collapse all files

262
docs/custom_integrations.html
View File

@@ -562,8 +562,12 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
</section>
<section id="usage" class="level3">
<h3 class="anchored" data-anchor-id="usage">Usage</h3>
<div class="sourceCode" id="cb3"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb3-2"><a href="#cb3-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<p><strong>NOTE</strong>: If you are training a VLM model, please use older version of Axolotl as upstream has applied a major VLM refactor, and our patches have not been updated yet.</p>
<div class="sourceCode" id="cb3"><pre class="sourceCode bash code-with-copy"><code class="sourceCode bash"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a><span class="fu">git</span> checkout 787880215b3ab32ccaf81c1b2e9588c6f3e6e764</span>
<span id="cb3-2"><a href="#cb3-2" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb3-3"><a href="#cb3-3" aria-hidden="true" tabindex="-1"></a><span class="ex">pip3</span> install <span class="at">--no-build-isolation</span> <span class="at">-e</span> .</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb4"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb4-2"><a href="#cb4-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
</section>
<section id="supported-models" class="level3">
<h3 class="anchored" data-anchor-id="supported-models">Supported Models</h3>
@@ -593,17 +597,17 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
</section>
<section id="citation" class="level3">
<h3 class="anchored" data-anchor-id="citation">Citation</h3>
<div class="sourceCode" id="cb4"><pre class="sourceCode bib code-with-copy"><code class="sourceCode bibtex"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a><span class="va">@article</span>{<span class="ot">wijmans2024cut</span>,</span>
<span id="cb4-2"><a href="#cb4-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">author</span> = {Erik Wijmans and</span>
<span id="cb4-3"><a href="#cb4-3" aria-hidden="true" tabindex="-1"></a> Brody Huval and</span>
<span id="cb4-4"><a href="#cb4-4" aria-hidden="true" tabindex="-1"></a> Alexander Hertzberg and</span>
<span id="cb4-5"><a href="#cb4-5" aria-hidden="true" tabindex="-1"></a> Vladlen Koltun and</span>
<span id="cb4-6"><a href="#cb4-6" aria-hidden="true" tabindex="-1"></a> Philipp Kr<span class="ch">\"</span>ahenb<span class="ch">\"</span>uhl},</span>
<span id="cb4-7"><a href="#cb4-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">title</span> = {Cut Your Losses in Large-Vocabulary Language Models},</span>
<span id="cb4-8"><a href="#cb4-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">journal</span> = {arXiv},</span>
<span id="cb4-9"><a href="#cb4-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">year</span> = {2024},</span>
<span id="cb4-10"><a href="#cb4-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">url</span> = {https://arxiv.org/abs/2411.09009},</span>
<span id="cb4-11"><a href="#cb4-11" aria-hidden="true" tabindex="-1"></a>}</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb5"><pre class="sourceCode bib code-with-copy"><code class="sourceCode bibtex"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a><span class="va">@article</span>{<span class="ot">wijmans2024cut</span>,</span>
<span id="cb5-2"><a href="#cb5-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">author</span> = {Erik Wijmans and</span>
<span id="cb5-3"><a href="#cb5-3" aria-hidden="true" tabindex="-1"></a> Brody Huval and</span>
<span id="cb5-4"><a href="#cb5-4" aria-hidden="true" tabindex="-1"></a> Alexander Hertzberg and</span>
<span id="cb5-5"><a href="#cb5-5" aria-hidden="true" tabindex="-1"></a> Vladlen Koltun and</span>
<span id="cb5-6"><a href="#cb5-6" aria-hidden="true" tabindex="-1"></a> Philipp Kr<span class="ch">\"</span>ahenb<span class="ch">\"</span>uhl},</span>
<span id="cb5-7"><a href="#cb5-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">title</span> = {Cut Your Losses in Large-Vocabulary Language Models},</span>
<span id="cb5-8"><a href="#cb5-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">journal</span> = {arXiv},</span>
<span id="cb5-9"><a href="#cb5-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">year</span> = {2024},</span>
<span id="cb5-10"><a href="#cb5-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">url</span> = {https://arxiv.org/abs/2411.09009},</span>
<span id="cb5-11"><a href="#cb5-11" aria-hidden="true" tabindex="-1"></a>}</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<p>Please see reference <a href="https://github.com/axolotl-ai-cloud/axolotl/tree/main/src/axolotl/integrations/cut_cross_entropy">here</a></p>
</section>
</section>
@@ -612,20 +616,20 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
<p>See https://github.com/ironjr/grokfast</p>
<section id="usage-1" class="level3">
<h3 class="anchored" data-anchor-id="usage-1">Usage</h3>
<div class="sourceCode" id="cb5"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb5-2"><a href="#cb5-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.grokfast.GrokfastPlugin</span></span>
<span id="cb5-3"><a href="#cb5-3" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb5-4"><a href="#cb5-4" aria-hidden="true" tabindex="-1"></a><span class="fu">grokfast_alpha</span><span class="kw">:</span><span class="at"> </span><span class="fl">2.0</span></span>
<span id="cb5-5"><a href="#cb5-5" aria-hidden="true" tabindex="-1"></a><span class="fu">grokfast_lamb</span><span class="kw">:</span><span class="at"> </span><span class="fl">0.98</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb6"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb6-1"><a href="#cb6-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb6-2"><a href="#cb6-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.grokfast.GrokfastPlugin</span></span>
<span id="cb6-3"><a href="#cb6-3" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb6-4"><a href="#cb6-4" aria-hidden="true" tabindex="-1"></a><span class="fu">grokfast_alpha</span><span class="kw">:</span><span class="at"> </span><span class="fl">2.0</span></span>
<span id="cb6-5"><a href="#cb6-5" aria-hidden="true" tabindex="-1"></a><span class="fu">grokfast_lamb</span><span class="kw">:</span><span class="at"> </span><span class="fl">0.98</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
</section>
<section id="citation-1" class="level3">
<h3 class="anchored" data-anchor-id="citation-1">Citation</h3>
<div class="sourceCode" id="cb6"><pre class="sourceCode bib code-with-copy"><code class="sourceCode bibtex"><span id="cb6-1"><a href="#cb6-1" aria-hidden="true" tabindex="-1"></a><span class="va">@article</span>{<span class="ot">lee2024grokfast</span>,</span>
<span id="cb6-2"><a href="#cb6-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">title</span>={{Grokfast}: Accelerated Grokking by Amplifying Slow Gradients},</span>
<span id="cb6-3"><a href="#cb6-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">author</span>={Lee, Jaerin and Kang, Bong Gyun and Kim, Kihoon and Lee, Kyoung Mu},</span>
<span id="cb6-4"><a href="#cb6-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">journal</span>={arXiv preprint arXiv:2405.20233},</span>
<span id="cb6-5"><a href="#cb6-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">year</span>={2024}</span>
<span id="cb6-6"><a href="#cb6-6" aria-hidden="true" tabindex="-1"></a>}</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb7"><pre class="sourceCode bib code-with-copy"><code class="sourceCode bibtex"><span id="cb7-1"><a href="#cb7-1" aria-hidden="true" tabindex="-1"></a><span class="va">@article</span>{<span class="ot">lee2024grokfast</span>,</span>
<span id="cb7-2"><a href="#cb7-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">title</span>={{Grokfast}: Accelerated Grokking by Amplifying Slow Gradients},</span>
<span id="cb7-3"><a href="#cb7-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">author</span>={Lee, Jaerin and Kang, Bong Gyun and Kim, Kihoon and Lee, Kyoung Mu},</span>
<span id="cb7-4"><a href="#cb7-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">journal</span>={arXiv preprint arXiv:2405.20233},</span>
<span id="cb7-5"><a href="#cb7-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">year</span>={2024}</span>
<span id="cb7-6"><a href="#cb7-6" aria-hidden="true" tabindex="-1"></a>}</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<p>Please see reference <a href="https://github.com/axolotl-ai-cloud/axolotl/tree/main/src/axolotl/integrations/grokfast">here</a></p>
</section>
</section>
@@ -633,21 +637,21 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
<h2 class="anchored" data-anchor-id="knowledge-distillation-kd">Knowledge Distillation (KD)</h2>
<section id="usage-2" class="level3">
<h3 class="anchored" data-anchor-id="usage-2">Usage</h3>
<div class="sourceCode" id="cb7"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb7-1"><a href="#cb7-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb7-2"><a href="#cb7-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> </span><span class="st">"axolotl.integrations.kd.KDPlugin"</span></span>
<span id="cb7-3"><a href="#cb7-3" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb7-4"><a href="#cb7-4" aria-hidden="true" tabindex="-1"></a><span class="fu">kd_trainer</span><span class="kw">:</span><span class="at"> </span><span class="ch">True</span></span>
<span id="cb7-5"><a href="#cb7-5" aria-hidden="true" tabindex="-1"></a><span class="fu">kd_ce_alpha</span><span class="kw">:</span><span class="at"> </span><span class="fl">0.1</span></span>
<span id="cb7-6"><a href="#cb7-6" aria-hidden="true" tabindex="-1"></a><span class="fu">kd_alpha</span><span class="kw">:</span><span class="at"> </span><span class="fl">0.9</span></span>
<span id="cb7-7"><a href="#cb7-7" aria-hidden="true" tabindex="-1"></a><span class="fu">kd_temperature</span><span class="kw">:</span><span class="at"> </span><span class="fl">1.0</span></span>
<span id="cb7-8"><a href="#cb7-8" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb7-9"><a href="#cb7-9" aria-hidden="true" tabindex="-1"></a><span class="fu">torch_compile</span><span class="kw">:</span><span class="at"> </span><span class="ch">True</span><span class="co"> # torch&gt;=2.5.1, recommended to reduce vram</span></span>
<span id="cb7-10"><a href="#cb7-10" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb7-11"><a href="#cb7-11" aria-hidden="true" tabindex="-1"></a><span class="fu">datasets</span><span class="kw">:</span></span>
<span id="cb7-12"><a href="#cb7-12" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> </span><span class="fu">path</span><span class="kw">:</span><span class="at"> ...</span></span>
<span id="cb7-13"><a href="#cb7-13" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">type</span><span class="kw">:</span><span class="at"> </span><span class="st">"axolotl.integrations.kd.chat_template"</span></span>
<span id="cb7-14"><a href="#cb7-14" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">field_messages</span><span class="kw">:</span><span class="at"> </span><span class="st">"messages_combined"</span></span>
<span id="cb7-15"><a href="#cb7-15" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">logprobs_field</span><span class="kw">:</span><span class="at"> </span><span class="st">"llm_text_generation_vllm_logprobs"</span><span class="co"> # for kd only, field of logprobs</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb8"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb8-1"><a href="#cb8-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb8-2"><a href="#cb8-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> </span><span class="st">"axolotl.integrations.kd.KDPlugin"</span></span>
<span id="cb8-3"><a href="#cb8-3" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb8-4"><a href="#cb8-4" aria-hidden="true" tabindex="-1"></a><span class="fu">kd_trainer</span><span class="kw">:</span><span class="at"> </span><span class="ch">True</span></span>
<span id="cb8-5"><a href="#cb8-5" aria-hidden="true" tabindex="-1"></a><span class="fu">kd_ce_alpha</span><span class="kw">:</span><span class="at"> </span><span class="fl">0.1</span></span>
<span id="cb8-6"><a href="#cb8-6" aria-hidden="true" tabindex="-1"></a><span class="fu">kd_alpha</span><span class="kw">:</span><span class="at"> </span><span class="fl">0.9</span></span>
<span id="cb8-7"><a href="#cb8-7" aria-hidden="true" tabindex="-1"></a><span class="fu">kd_temperature</span><span class="kw">:</span><span class="at"> </span><span class="fl">1.0</span></span>
<span id="cb8-8"><a href="#cb8-8" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb8-9"><a href="#cb8-9" aria-hidden="true" tabindex="-1"></a><span class="fu">torch_compile</span><span class="kw">:</span><span class="at"> </span><span class="ch">True</span><span class="co"> # torch&gt;=2.5.1, recommended to reduce vram</span></span>
<span id="cb8-10"><a href="#cb8-10" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb8-11"><a href="#cb8-11" aria-hidden="true" tabindex="-1"></a><span class="fu">datasets</span><span class="kw">:</span></span>
<span id="cb8-12"><a href="#cb8-12" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> </span><span class="fu">path</span><span class="kw">:</span><span class="at"> ...</span></span>
<span id="cb8-13"><a href="#cb8-13" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">type</span><span class="kw">:</span><span class="at"> </span><span class="st">"axolotl.integrations.kd.chat_template"</span></span>
<span id="cb8-14"><a href="#cb8-14" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">field_messages</span><span class="kw">:</span><span class="at"> </span><span class="st">"messages_combined"</span></span>
<span id="cb8-15"><a href="#cb8-15" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">logprobs_field</span><span class="kw">:</span><span class="at"> </span><span class="st">"llm_text_generation_vllm_logprobs"</span><span class="co"> # for kd only, field of logprobs</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<p>An example dataset can be found at <a href="https://huggingface.co/datasets/axolotl-ai-co/evolkit-logprobs-pipeline-75k-v2-sample"><code>axolotl-ai-co/evolkit-logprobs-pipeline-75k-v2-sample</code></a></p>
<p>Please see reference <a href="https://github.com/axolotl-ai-cloud/axolotl/tree/main/src/axolotl/integrations/kd">here</a></p>
</section>
@@ -663,13 +667,13 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
<p>See https://github.com/linkedin/Liger-Kernel</p>
<section id="usage-3" class="level3">
<h3 class="anchored" data-anchor-id="usage-3">Usage</h3>
<div class="sourceCode" id="cb8"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb8-1"><a href="#cb8-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb8-2"><a href="#cb8-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.liger.LigerPlugin</span></span>
<span id="cb8-3"><a href="#cb8-3" aria-hidden="true" tabindex="-1"></a><span class="fu">liger_rope</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span>
<span id="cb8-4"><a href="#cb8-4" aria-hidden="true" tabindex="-1"></a><span class="fu">liger_rms_norm</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span>
<span id="cb8-5"><a href="#cb8-5" aria-hidden="true" tabindex="-1"></a><span class="fu">liger_glu_activation</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span>
<span id="cb8-6"><a href="#cb8-6" aria-hidden="true" tabindex="-1"></a><span class="fu">liger_layer_norm</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span>
<span id="cb8-7"><a href="#cb8-7" aria-hidden="true" tabindex="-1"></a><span class="fu">liger_fused_linear_cross_entropy</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb9"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb9-2"><a href="#cb9-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.liger.LigerPlugin</span></span>
<span id="cb9-3"><a href="#cb9-3" aria-hidden="true" tabindex="-1"></a><span class="fu">liger_rope</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span>
<span id="cb9-4"><a href="#cb9-4" aria-hidden="true" tabindex="-1"></a><span class="fu">liger_rms_norm</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span>
<span id="cb9-5"><a href="#cb9-5" aria-hidden="true" tabindex="-1"></a><span class="fu">liger_glu_activation</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span>
<span id="cb9-6"><a href="#cb9-6" aria-hidden="true" tabindex="-1"></a><span class="fu">liger_layer_norm</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span>
<span id="cb9-7"><a href="#cb9-7" aria-hidden="true" tabindex="-1"></a><span class="fu">liger_fused_linear_cross_entropy</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
</section>
<section id="supported-models-1" class="level3">
<h3 class="anchored" data-anchor-id="supported-models-1">Supported Models</h3>
@@ -695,16 +699,16 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
</section>
<section id="citation-2" class="level3">
<h3 class="anchored" data-anchor-id="citation-2">Citation</h3>
<div class="sourceCode" id="cb9"><pre class="sourceCode bib code-with-copy"><code class="sourceCode bibtex"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a><span class="va">@article</span>{<span class="ot">hsu2024ligerkernelefficienttriton</span>,</span>
<span id="cb9-2"><a href="#cb9-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">title</span>={Liger Kernel: Efficient Triton Kernels for LLM Training},</span>
<span id="cb9-3"><a href="#cb9-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">author</span>={Pin-Lun Hsu and Yun Dai and Vignesh Kothapalli and Qingquan Song and Shao Tang and Siyu Zhu and Steven Shimizu and Shivam Sahni and Haowen Ning and Yanning Chen},</span>
<span id="cb9-4"><a href="#cb9-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">year</span>={2024},</span>
<span id="cb9-5"><a href="#cb9-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">eprint</span>={2410.10989},</span>
<span id="cb9-6"><a href="#cb9-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">archivePrefix</span>={arXiv},</span>
<span id="cb9-7"><a href="#cb9-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">primaryClass</span>={cs.LG},</span>
<span id="cb9-8"><a href="#cb9-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">url</span>={https://arxiv.org/abs/2410.10989},</span>
<span id="cb9-9"><a href="#cb9-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">journal</span>={arXiv preprint arXiv:2410.10989},</span>
<span id="cb9-10"><a href="#cb9-10" aria-hidden="true" tabindex="-1"></a>}</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb10"><pre class="sourceCode bib code-with-copy"><code class="sourceCode bibtex"><span id="cb10-1"><a href="#cb10-1" aria-hidden="true" tabindex="-1"></a><span class="va">@article</span>{<span class="ot">hsu2024ligerkernelefficienttriton</span>,</span>
<span id="cb10-2"><a href="#cb10-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">title</span>={Liger Kernel: Efficient Triton Kernels for LLM Training},</span>
<span id="cb10-3"><a href="#cb10-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">author</span>={Pin-Lun Hsu and Yun Dai and Vignesh Kothapalli and Qingquan Song and Shao Tang and Siyu Zhu and Steven Shimizu and Shivam Sahni and Haowen Ning and Yanning Chen},</span>
<span id="cb10-4"><a href="#cb10-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">year</span>={2024},</span>
<span id="cb10-5"><a href="#cb10-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">eprint</span>={2410.10989},</span>
<span id="cb10-6"><a href="#cb10-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">archivePrefix</span>={arXiv},</span>
<span id="cb10-7"><a href="#cb10-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">primaryClass</span>={cs.LG},</span>
<span id="cb10-8"><a href="#cb10-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">url</span>={https://arxiv.org/abs/2410.10989},</span>
<span id="cb10-9"><a href="#cb10-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">journal</span>={arXiv preprint arXiv:2410.10989},</span>
<span id="cb10-10"><a href="#cb10-10" aria-hidden="true" tabindex="-1"></a>}</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<p>Please see reference <a href="https://github.com/axolotl-ai-cloud/axolotl/tree/main/src/axolotl/integrations/liger">here</a></p>
</section>
</section>
@@ -714,29 +718,29 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
<p>See https://github.com/EleutherAI/lm-evaluation-harness</p>
<section id="usage-4" class="level3">
<h3 class="anchored" data-anchor-id="usage-4">Usage</h3>
<div class="sourceCode" id="cb10"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb10-1"><a href="#cb10-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb10-2"><a href="#cb10-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.lm_eval.LMEvalPlugin</span></span>
<span id="cb10-3"><a href="#cb10-3" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb10-4"><a href="#cb10-4" aria-hidden="true" tabindex="-1"></a><span class="fu">lm_eval_tasks</span><span class="kw">:</span></span>
<span id="cb10-5"><a href="#cb10-5" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> gsm8k</span></span>
<span id="cb10-6"><a href="#cb10-6" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> hellaswag</span></span>
<span id="cb10-7"><a href="#cb10-7" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> arc_easy</span></span>
<span id="cb10-8"><a href="#cb10-8" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb10-9"><a href="#cb10-9" aria-hidden="true" tabindex="-1"></a><span class="fu">lm_eval_batch_size</span><span class="kw">:</span><span class="co"> # Batch size for evaluation</span></span>
<span id="cb10-10"><a href="#cb10-10" aria-hidden="true" tabindex="-1"></a><span class="fu">output_dir</span><span class="kw">:</span><span class="co"> # Directory to save evaluation results</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb11"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb11-1"><a href="#cb11-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb11-2"><a href="#cb11-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.lm_eval.LMEvalPlugin</span></span>
<span id="cb11-3"><a href="#cb11-3" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb11-4"><a href="#cb11-4" aria-hidden="true" tabindex="-1"></a><span class="fu">lm_eval_tasks</span><span class="kw">:</span></span>
<span id="cb11-5"><a href="#cb11-5" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> gsm8k</span></span>
<span id="cb11-6"><a href="#cb11-6" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> hellaswag</span></span>
<span id="cb11-7"><a href="#cb11-7" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> arc_easy</span></span>
<span id="cb11-8"><a href="#cb11-8" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb11-9"><a href="#cb11-9" aria-hidden="true" tabindex="-1"></a><span class="fu">lm_eval_batch_size</span><span class="kw">:</span><span class="co"> # Batch size for evaluation</span></span>
<span id="cb11-10"><a href="#cb11-10" aria-hidden="true" tabindex="-1"></a><span class="fu">output_dir</span><span class="kw">:</span><span class="co"> # Directory to save evaluation results</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
</section>
<section id="citation-3" class="level3">
<h3 class="anchored" data-anchor-id="citation-3">Citation</h3>
<div class="sourceCode" id="cb11"><pre class="sourceCode bib code-with-copy"><code class="sourceCode bibtex"><span id="cb11-1"><a href="#cb11-1" aria-hidden="true" tabindex="-1"></a><span class="va">@misc</span>{<span class="ot">eval</span>-<span class="ot">harness</span>,</span>
<span id="cb11-2"><a href="#cb11-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">author</span> = {Gao, Leo and Tow, Jonathan and Abbasi, Baber and Biderman, Stella and Black, Sid and DiPofi, Anthony and Foster, Charles and Golding, Laurence and Hsu, Jeffrey and Le Noac'h, Alain and Li, Haonan and McDonell, Kyle and Muennighoff, Niklas and Ociepa, Chris and Phang, Jason and Reynolds, Laria and Schoelkopf, Hailey and Skowron, Aviya and Sutawika, Lintang and Tang, Eric and Thite, Anish and Wang, Ben and Wang, Kevin and Zou, Andy},</span>
<span id="cb11-3"><a href="#cb11-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">title</span> = {A framework for few-shot language model evaluation},</span>
<span id="cb11-4"><a href="#cb11-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">month</span> = 07,</span>
<span id="cb11-5"><a href="#cb11-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">year</span> = 2024,</span>
<span id="cb11-6"><a href="#cb11-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">publisher</span> = {Zenodo},</span>
<span id="cb11-7"><a href="#cb11-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">version</span> = {v0.4.3},</span>
<span id="cb11-8"><a href="#cb11-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">doi</span> = {10.5281/zenodo.12608602},</span>
<span id="cb11-9"><a href="#cb11-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">url</span> = {https://zenodo.org/records/12608602}</span>
<span id="cb11-10"><a href="#cb11-10" aria-hidden="true" tabindex="-1"></a>}</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb12"><pre class="sourceCode bib code-with-copy"><code class="sourceCode bibtex"><span id="cb12-1"><a href="#cb12-1" aria-hidden="true" tabindex="-1"></a><span class="va">@misc</span>{<span class="ot">eval</span>-<span class="ot">harness</span>,</span>
<span id="cb12-2"><a href="#cb12-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">author</span> = {Gao, Leo and Tow, Jonathan and Abbasi, Baber and Biderman, Stella and Black, Sid and DiPofi, Anthony and Foster, Charles and Golding, Laurence and Hsu, Jeffrey and Le Noac'h, Alain and Li, Haonan and McDonell, Kyle and Muennighoff, Niklas and Ociepa, Chris and Phang, Jason and Reynolds, Laria and Schoelkopf, Hailey and Skowron, Aviya and Sutawika, Lintang and Tang, Eric and Thite, Anish and Wang, Ben and Wang, Kevin and Zou, Andy},</span>
<span id="cb12-3"><a href="#cb12-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">title</span> = {A framework for few-shot language model evaluation},</span>
<span id="cb12-4"><a href="#cb12-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">month</span> = 07,</span>
<span id="cb12-5"><a href="#cb12-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">year</span> = 2024,</span>
<span id="cb12-6"><a href="#cb12-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">publisher</span> = {Zenodo},</span>
<span id="cb12-7"><a href="#cb12-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">version</span> = {v0.4.3},</span>
<span id="cb12-8"><a href="#cb12-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">doi</span> = {10.5281/zenodo.12608602},</span>
<span id="cb12-9"><a href="#cb12-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">url</span> = {https://zenodo.org/records/12608602}</span>
<span id="cb12-10"><a href="#cb12-10" aria-hidden="true" tabindex="-1"></a>}</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<p>Please see reference <a href="https://github.com/axolotl-ai-cloud/axolotl/tree/main/src/axolotl/integrations/lm_eval">here</a></p>
</section>
</section>
@@ -752,23 +756,23 @@ By identifying the top n% of layers with the highest SNR, you can optimize train
</section>
<section id="usage-5" class="level3">
<h3 class="anchored" data-anchor-id="usage-5">Usage</h3>
<div class="sourceCode" id="cb12"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb12-1"><a href="#cb12-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb12-2"><a href="#cb12-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.spectrum.SpectrumPlugin</span></span>
<span id="cb12-3"><a href="#cb12-3" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb12-4"><a href="#cb12-4" aria-hidden="true" tabindex="-1"></a><span class="fu">spectrum_top_fraction</span><span class="kw">:</span><span class="at"> </span><span class="fl">0.5</span></span>
<span id="cb12-5"><a href="#cb12-5" aria-hidden="true" tabindex="-1"></a><span class="fu">spectrum_model_name</span><span class="kw">:</span><span class="at"> meta-llama/Meta-Llama-3.1-8B</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb13"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb13-1"><a href="#cb13-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb13-2"><a href="#cb13-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.spectrum.SpectrumPlugin</span></span>
<span id="cb13-3"><a href="#cb13-3" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb13-4"><a href="#cb13-4" aria-hidden="true" tabindex="-1"></a><span class="fu">spectrum_top_fraction</span><span class="kw">:</span><span class="at"> </span><span class="fl">0.5</span></span>
<span id="cb13-5"><a href="#cb13-5" aria-hidden="true" tabindex="-1"></a><span class="fu">spectrum_model_name</span><span class="kw">:</span><span class="at"> meta-llama/Meta-Llama-3.1-8B</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
</section>
<section id="citation-4" class="level3">
<h3 class="anchored" data-anchor-id="citation-4">Citation</h3>
<div class="sourceCode" id="cb13"><pre class="sourceCode bib code-with-copy"><code class="sourceCode bibtex"><span id="cb13-1"><a href="#cb13-1" aria-hidden="true" tabindex="-1"></a><span class="va">@misc</span>{<span class="ot">hartford2024spectrumtargetedtrainingsignal</span>,</span>
<span id="cb13-2"><a href="#cb13-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">title</span>={Spectrum: Targeted Training on Signal to Noise Ratio},</span>
<span id="cb13-3"><a href="#cb13-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">author</span>={Eric Hartford and Lucas Atkins and Fernando Fernandes Neto and David Golchinfar},</span>
<span id="cb13-4"><a href="#cb13-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">year</span>={2024},</span>
<span id="cb13-5"><a href="#cb13-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">eprint</span>={2406.06623},</span>
<span id="cb13-6"><a href="#cb13-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">archivePrefix</span>={arXiv},</span>
<span id="cb13-7"><a href="#cb13-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">primaryClass</span>={cs.LG},</span>
<span id="cb13-8"><a href="#cb13-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">url</span>={https://arxiv.org/abs/2406.06623},</span>
<span id="cb13-9"><a href="#cb13-9" aria-hidden="true" tabindex="-1"></a>}</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb14"><pre class="sourceCode bib code-with-copy"><code class="sourceCode bibtex"><span id="cb14-1"><a href="#cb14-1" aria-hidden="true" tabindex="-1"></a><span class="va">@misc</span>{<span class="ot">hartford2024spectrumtargetedtrainingsignal</span>,</span>
<span id="cb14-2"><a href="#cb14-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">title</span>={Spectrum: Targeted Training on Signal to Noise Ratio},</span>
<span id="cb14-3"><a href="#cb14-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">author</span>={Eric Hartford and Lucas Atkins and Fernando Fernandes Neto and David Golchinfar},</span>
<span id="cb14-4"><a href="#cb14-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">year</span>={2024},</span>
<span id="cb14-5"><a href="#cb14-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">eprint</span>={2406.06623},</span>
<span id="cb14-6"><a href="#cb14-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">archivePrefix</span>={arXiv},</span>
<span id="cb14-7"><a href="#cb14-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">primaryClass</span>={cs.LG},</span>
<span id="cb14-8"><a href="#cb14-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">url</span>={https://arxiv.org/abs/2406.06623},</span>
<span id="cb14-9"><a href="#cb14-9" aria-hidden="true" tabindex="-1"></a>}</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<p>Please see reference <a href="https://github.com/axolotl-ai-cloud/axolotl/tree/main/src/axolotl/integrations/spectrum">here</a></p>
</section>
</section>
@@ -782,7 +786,7 @@ By identifying the top n% of layers with the highest SNR, you can optimize train
<h3 class="anchored" data-anchor-id="requirements-1">Requirements</h3>
<ul>
<li><p>Axolotl with <code>llmcompressor</code> extras:</p>
<div class="sourceCode" id="cb14"><pre class="sourceCode bash code-with-copy"><code class="sourceCode bash"><span id="cb14-1"><a href="#cb14-1" aria-hidden="true" tabindex="-1"></a><span class="ex">pip</span> install <span class="st">"axolotl[llmcompressor]"</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div></li>
<div class="sourceCode" id="cb15"><pre class="sourceCode bash code-with-copy"><code class="sourceCode bash"><span id="cb15-1"><a href="#cb15-1" aria-hidden="true" tabindex="-1"></a><span class="ex">pip</span> install <span class="st">"axolotl[llmcompressor]"</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div></li>
<li><p>Requires <code>llmcompressor &gt;= 0.5.1</code></p></li>
</ul>
<p>This will install all necessary dependencies to fine-tune sparsified models using the integration.</p>
@@ -791,25 +795,25 @@ By identifying the top n% of layers with the highest SNR, you can optimize train
<section id="usage-6" class="level3">
<h3 class="anchored" data-anchor-id="usage-6">Usage</h3>
<p>To enable sparse fine-tuning with this integration, include the plugin in your Axolotl config:</p>
<div class="sourceCode" id="cb15"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb15-1"><a href="#cb15-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb15-2"><a href="#cb15-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.llm_compressor.LLMCompressorPlugin</span></span>
<span id="cb15-3"><a href="#cb15-3" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb15-4"><a href="#cb15-4" aria-hidden="true" tabindex="-1"></a><span class="fu">llmcompressor</span><span class="kw">:</span></span>
<span id="cb15-5"><a href="#cb15-5" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">recipe</span><span class="kw">:</span></span>
<span id="cb15-6"><a href="#cb15-6" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">finetuning_stage</span><span class="kw">:</span></span>
<span id="cb15-7"><a href="#cb15-7" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">finetuning_modifiers</span><span class="kw">:</span></span>
<span id="cb15-8"><a href="#cb15-8" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">ConstantPruningModifier</span><span class="kw">:</span></span>
<span id="cb15-9"><a href="#cb15-9" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">targets</span><span class="kw">:</span><span class="at"> </span><span class="kw">[</span></span>
<span id="cb15-10"><a href="#cb15-10" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*q_proj.weight'</span><span class="kw">,</span></span>
<span id="cb15-11"><a href="#cb15-11" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*k_proj.weight'</span><span class="kw">,</span></span>
<span id="cb15-12"><a href="#cb15-12" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*v_proj.weight'</span><span class="kw">,</span></span>
<span id="cb15-13"><a href="#cb15-13" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*o_proj.weight'</span><span class="kw">,</span></span>
<span id="cb15-14"><a href="#cb15-14" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*gate_proj.weight'</span><span class="kw">,</span></span>
<span id="cb15-15"><a href="#cb15-15" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*up_proj.weight'</span><span class="kw">,</span></span>
<span id="cb15-16"><a href="#cb15-16" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*down_proj.weight'</span><span class="kw">,</span></span>
<span id="cb15-17"><a href="#cb15-17" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">]</span></span>
<span id="cb15-18"><a href="#cb15-18" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">start</span><span class="kw">:</span><span class="at"> </span><span class="dv">0</span></span>
<span id="cb15-19"><a href="#cb15-19" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">save_compressed</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb16"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb16-1"><a href="#cb16-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb16-2"><a href="#cb16-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.llm_compressor.LLMCompressorPlugin</span></span>
<span id="cb16-3"><a href="#cb16-3" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb16-4"><a href="#cb16-4" aria-hidden="true" tabindex="-1"></a><span class="fu">llmcompressor</span><span class="kw">:</span></span>
<span id="cb16-5"><a href="#cb16-5" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">recipe</span><span class="kw">:</span></span>
<span id="cb16-6"><a href="#cb16-6" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">finetuning_stage</span><span class="kw">:</span></span>
<span id="cb16-7"><a href="#cb16-7" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">finetuning_modifiers</span><span class="kw">:</span></span>
<span id="cb16-8"><a href="#cb16-8" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">ConstantPruningModifier</span><span class="kw">:</span></span>
<span id="cb16-9"><a href="#cb16-9" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">targets</span><span class="kw">:</span><span class="at"> </span><span class="kw">[</span></span>
<span id="cb16-10"><a href="#cb16-10" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*q_proj.weight'</span><span class="kw">,</span></span>
<span id="cb16-11"><a href="#cb16-11" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*k_proj.weight'</span><span class="kw">,</span></span>
<span id="cb16-12"><a href="#cb16-12" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*v_proj.weight'</span><span class="kw">,</span></span>
<span id="cb16-13"><a href="#cb16-13" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*o_proj.weight'</span><span class="kw">,</span></span>
<span id="cb16-14"><a href="#cb16-14" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*gate_proj.weight'</span><span class="kw">,</span></span>
<span id="cb16-15"><a href="#cb16-15" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*up_proj.weight'</span><span class="kw">,</span></span>
<span id="cb16-16"><a href="#cb16-16" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="st">'re:.*down_proj.weight'</span><span class="kw">,</span></span>
<span id="cb16-17"><a href="#cb16-17" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">]</span></span>
<span id="cb16-18"><a href="#cb16-18" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">start</span><span class="kw">:</span><span class="at"> </span><span class="dv">0</span></span>
<span id="cb16-19"><a href="#cb16-19" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">save_compressed</span><span class="kw">:</span><span class="at"> </span><span class="ch">true</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<p>This plugin <strong>does not apply pruning or sparsification itself</strong> — it is intended for <strong>fine-tuning models that have already been sparsified</strong>.</p>
<p>Pre-sparsified checkpoints can be:
- Generated using <a href="https://github.com/vllm-project/llm-compressor">LLMCompressor</a>
@@ -836,22 +840,22 @@ By identifying the top n% of layers with the highest SNR, you can optimize train
<p>After fine-tuning your sparse model, you can leverage vLLM for efficient inference.
You can also use LLMCompressor to apply additional quantization to your fine-tuned
sparse model before inference for even greater performance benefits.:</p>
<div class="sourceCode" id="cb16"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb16-1"><a href="#cb16-1" aria-hidden="true" tabindex="-1"></a><span class="im">from</span> vllm <span class="im">import</span> LLM, SamplingParams</span>
<span id="cb16-2"><a href="#cb16-2" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb16-3"><a href="#cb16-3" aria-hidden="true" tabindex="-1"></a>prompts <span class="op">=</span> [</span>
<span id="cb16-4"><a href="#cb16-4" aria-hidden="true" tabindex="-1"></a> <span class="st">"Hello, my name is"</span>,</span>
<span id="cb16-5"><a href="#cb16-5" aria-hidden="true" tabindex="-1"></a> <span class="st">"The president of the United States is"</span>,</span>
<span id="cb16-6"><a href="#cb16-6" aria-hidden="true" tabindex="-1"></a> <span class="st">"The capital of France is"</span>,</span>
<span id="cb16-7"><a href="#cb16-7" aria-hidden="true" tabindex="-1"></a> <span class="st">"The future of AI is"</span>,</span>
<span id="cb16-8"><a href="#cb16-8" aria-hidden="true" tabindex="-1"></a>]</span>
<span id="cb16-9"><a href="#cb16-9" aria-hidden="true" tabindex="-1"></a>sampling_params <span class="op">=</span> SamplingParams(temperature<span class="op">=</span><span class="fl">0.8</span>, top_p<span class="op">=</span><span class="fl">0.95</span>)</span>
<span id="cb16-10"><a href="#cb16-10" aria-hidden="true" tabindex="-1"></a>llm <span class="op">=</span> LLM(<span class="st">"path/to/your/sparse/model"</span>)</span>
<span id="cb16-11"><a href="#cb16-11" aria-hidden="true" tabindex="-1"></a>outputs <span class="op">=</span> llm.generate(prompts, sampling_params)</span>
<span id="cb16-12"><a href="#cb16-12" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb16-13"><a href="#cb16-13" aria-hidden="true" tabindex="-1"></a><span class="cf">for</span> output <span class="kw">in</span> outputs:</span>
<span id="cb16-14"><a href="#cb16-14" aria-hidden="true" tabindex="-1"></a> prompt <span class="op">=</span> output.prompt</span>
<span id="cb16-15"><a href="#cb16-15" aria-hidden="true" tabindex="-1"></a> generated_text <span class="op">=</span> output.outputs[<span class="dv">0</span>].text</span>
<span id="cb16-16"><a href="#cb16-16" aria-hidden="true" tabindex="-1"></a> <span class="bu">print</span>(<span class="ss">f"Prompt: </span><span class="sc">{</span>prompt<span class="sc">!r}</span><span class="ss">, Generated text: </span><span class="sc">{</span>generated_text<span class="sc">!r}</span><span class="ss">"</span>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb17"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb17-1"><a href="#cb17-1" aria-hidden="true" tabindex="-1"></a><span class="im">from</span> vllm <span class="im">import</span> LLM, SamplingParams</span>
<span id="cb17-2"><a href="#cb17-2" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb17-3"><a href="#cb17-3" aria-hidden="true" tabindex="-1"></a>prompts <span class="op">=</span> [</span>
<span id="cb17-4"><a href="#cb17-4" aria-hidden="true" tabindex="-1"></a> <span class="st">"Hello, my name is"</span>,</span>
<span id="cb17-5"><a href="#cb17-5" aria-hidden="true" tabindex="-1"></a> <span class="st">"The president of the United States is"</span>,</span>
<span id="cb17-6"><a href="#cb17-6" aria-hidden="true" tabindex="-1"></a> <span class="st">"The capital of France is"</span>,</span>
<span id="cb17-7"><a href="#cb17-7" aria-hidden="true" tabindex="-1"></a> <span class="st">"The future of AI is"</span>,</span>
<span id="cb17-8"><a href="#cb17-8" aria-hidden="true" tabindex="-1"></a>]</span>
<span id="cb17-9"><a href="#cb17-9" aria-hidden="true" tabindex="-1"></a>sampling_params <span class="op">=</span> SamplingParams(temperature<span class="op">=</span><span class="fl">0.8</span>, top_p<span class="op">=</span><span class="fl">0.95</span>)</span>
<span id="cb17-10"><a href="#cb17-10" aria-hidden="true" tabindex="-1"></a>llm <span class="op">=</span> LLM(<span class="st">"path/to/your/sparse/model"</span>)</span>
<span id="cb17-11"><a href="#cb17-11" aria-hidden="true" tabindex="-1"></a>outputs <span class="op">=</span> llm.generate(prompts, sampling_params)</span>
<span id="cb17-12"><a href="#cb17-12" aria-hidden="true" tabindex="-1"></a></span>
<span id="cb17-13"><a href="#cb17-13" aria-hidden="true" tabindex="-1"></a><span class="cf">for</span> output <span class="kw">in</span> outputs:</span>
<span id="cb17-14"><a href="#cb17-14" aria-hidden="true" tabindex="-1"></a> prompt <span class="op">=</span> output.prompt</span>
<span id="cb17-15"><a href="#cb17-15" aria-hidden="true" tabindex="-1"></a> generated_text <span class="op">=</span> output.outputs[<span class="dv">0</span>].text</span>
<span id="cb17-16"><a href="#cb17-16" aria-hidden="true" tabindex="-1"></a> <span class="bu">print</span>(<span class="ss">f"Prompt: </span><span class="sc">{</span>prompt<span class="sc">!r}</span><span class="ss">, Generated text: </span><span class="sc">{</span>generated_text<span class="sc">!r}</span><span class="ss">"</span>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<p>For more details on vLLM’s capabilities and advanced configuration options, see the <a href="https://docs.vllm.ai/">official vLLM documentation</a>.</p>
</section>
<section id="learn-more" class="level3">
@@ -901,10 +905,10 @@ Warning
</div>
<div class="callout-body-container callout-body">
<p>If you could not load your integration, please ensure you are pip installing in editable mode.</p>
<div class="sourceCode" id="cb17"><pre class="sourceCode bash code-with-copy"><code class="sourceCode bash"><span id="cb17-1"><a href="#cb17-1" aria-hidden="true" tabindex="-1"></a><span class="ex">pip</span> install <span class="at">-e</span> .</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb18"><pre class="sourceCode bash code-with-copy"><code class="sourceCode bash"><span id="cb18-1"><a href="#cb18-1" aria-hidden="true" tabindex="-1"></a><span class="ex">pip</span> install <span class="at">-e</span> .</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<p>and correctly spelled the integration name in the config file.</p>
<div class="sourceCode" id="cb18"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb18-1"><a href="#cb18-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb18-2"><a href="#cb18-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.your_integration_name.YourIntegrationPlugin</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="sourceCode" id="cb19"><pre class="sourceCode yaml code-with-copy"><code class="sourceCode yaml"><span id="cb19-1"><a href="#cb19-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plugins</span><span class="kw">:</span></span>
<span id="cb19-2"><a href="#cb19-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="kw">-</span><span class="at"> axolotl.integrations.your_integration_name.YourIntegrationPlugin</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
</div>
</div>
<div class="callout callout-style-default callout-note callout-titled">