Update README.md

README.md

@@ -141,7 +141,7 @@ library_name: transformers
                         </p>
                         <p>
                             The total dataset size is around 1.7B tokens, 1.1B
-                            of which are creative and 600M are technical.
+                            of which is creative and 600M being technical.
                         </p>
                         <p>
                             V1.3.0 introduces multilingual capabilities with
@@ -174,24 +174,20 @@ REPETITION_PENALTY: 1.04</pre
                             The model uses the following format I'll refer to as
                             "DanChat-2":
                         </p>
-                        <h4>Why not ChatML?</h4>
-                        <p>
-                            ChatML is a widely used and standardized format for
-                            LLMs but it has some limitations, using standard
-                            tokens as turn ownership indicators can impart
-                            biases to the model. DanChat-2 uses unique special
-                            tokens for each role, which helps to reduce these
-                            biases and allow the model to more readily adapt to
-                            different roles and tasks. Yes, It is possible to
-                            achieve a similar effect with ChatML, but the
-                            technique to do so would be nonstandard to the
-                            ChatML format and for users who do not use standard
-                            "assistant" and "user" roles, it would fall apart
-                            entirely.
-                        </p>
                         <pre class="code-block">
 <|system|>system prompt<|endoftext|><|user|>Hi there!<|endoftext|><|assistant|>Hey, how can I help?<|endoftext|></pre
                         >
+                        <h3>Why not ChatML?</h3>
+                        <p>
+                            While ChatML is a standard format for LLMs, it has
+                            limitations. DanChat-2 uses unique special tokens
+                            for each role, reducing biases and helping the model
+                            adapt to different tasks better than ChatML's
+                            standard tokens. Though similar effects could be
+                            achieved with modified ChatML, such approaches would
+                            be non-standard and wouldn't work for users with
+                            custom roles beyond "assistant" and "user".
+                        </p>
                         <h3>SillyTavern Template</h3>
                         <p>
                             <a
@@ -203,6 +199,20 @@ REPETITION_PENALTY: 1.04</pre
                                 Download Master JSON
                             </a>
                         </p>
+                        <h3>Inference Provider</h3>
+                        This model and others are available from ⚡Mancer AI for
+                        those interested in high quality inference without
+                        owning or renting expensive hardware.
+                        <p class="mancer-button-container">
+                            <a
+                                href="https://mancer.tech/"
+                                target="_blank"
+                                rel="noopener noreferrer"
+                                class="mancer-button"
+                            >
+                                ⚡mancer
+                            </a>
+                        </p>
                         <h3>Training Process</h3>
                         <p>
                             The model was trained using Axolotl on 8x H100 GPUs
@@ -429,25 +439,29 @@ REPETITION_PENALTY: 1.04</pre
             color: #e49b3e;
             overflow-x: auto; /* Added to enable horizontal scrolling */
         }
-        .sponsor-image {
-            width: 360px;
-            height: auto;
-            border: 2px solid #e49b3e;
-            border-radius: 10px;
-            filter: brightness(0.9) sepia(0.2);
-            transition: all 0.3s ease;
+        .mancer-button-container {
+            text-align: left;
+            margin: 1em 0;
         }
-        .sponsor-image-small {
-            width: 180px;
-            height: auto;
+        .mancer-button {
+            display: inline-block;
+            background: #1a1a1a;
+            color: #e49b3e;
+            padding: 10px 20px;
             border: 2px solid #e49b3e;
             border-radius: 5px;
-            filter: brightness(0.9) sepia(0.2);
+            font-family: "VT323", monospace;
+            font-size: 18px;
+            text-decoration: none !important; /* Ensures no underline */
+            text-shadow: 0 0 2px #e49b3e;
+            box-shadow: 0 0 10px rgba(228, 155, 62, 0.3);
             transition: all 0.3s ease;
         }
-        .sponsor-image:hover {
-            filter: brightness(1) sepia(0);
-            box-shadow: 0 0 10px rgba(228, 155, 62, 0.5);
+        .mancer-button:hover {
+            background: #2a2a2a;
+            box-shadow: 0 0 15px rgba(228, 155, 62, 0.5);
+            text-shadow: 0 0 4px #e49b3e;
+            text-decoration: none !important; /* Ensures no underline on hover */
         }
     </style>
 </html>