Upload model

README.md

@@ -0,0 +1,41 @@
+---
+tags:
+- t5
+- adapter-transformers
+- adapterhub:sum/xsum
+---
+
+# Adapter `ilhami/my-awesome-adapter` for t5-small
+
+An [adapter](https://adapterhub.ml) for the `t5-small` model that was trained on the [sum/xsum](https://adapterhub.ml/explore/sum/xsum/) dataset and includes a prediction head for seq2seq lm.
+
+This adapter was created for usage with the **[Adapters](https://github.com/Adapter-Hub/adapters)** library.
+
+## Usage
+
+First, install `adapters`:
+
+```
+pip install -U adapters
+```
+
+Now, the adapter can be loaded and activated like this:
+
+```python
+from adapters import AutoAdapterModel
+
+model = AutoAdapterModel.from_pretrained("t5-small")
+adapter_name = model.load_adapter("ilhami/my-awesome-adapter", source="hf", set_active=True)
+```
+
+## Architecture & Training
+
+<!-- Add some description here -->
+
+## Evaluation results
+
+<!-- Add some description here -->
+
+## Citation
+
+<!-- Add some description here -->

adapter_config.json

@@ -0,0 +1,75 @@
+{
+  "config": {
+    "architecture": "union",
+    "configs": [
+      {
+        "alpha": 8,
+        "architecture": "lora",
+        "attn_matrices": [
+          "q",
+          "v"
+        ],
+        "composition_mode": "add",
+        "dropout": 0.0,
+        "init_weights": "lora",
+        "intermediate_lora": false,
+        "leave_out": [],
+        "output_lora": false,
+        "r": 8,
+        "selfattn_lora": true,
+        "use_gating": true
+      },
+      {
+        "architecture": "prefix_tuning",
+        "bottleneck_size": 512,
+        "cross_prefix": true,
+        "dropout": 0.0,
+        "encoder_prefix": true,
+        "flat": false,
+        "leave_out": [],
+        "non_linearity": "tanh",
+        "prefix_length": 10,
+        "shared_gating": true,
+        "use_gating": true
+      },
+      {
+        "adapter_residual_before_ln": false,
+        "cross_adapter": false,
+        "factorized_phm_W": true,
+        "factorized_phm_rule": false,
+        "hypercomplex_nonlinearity": "glorot-uniform",
+        "init_weights": "bert",
+        "inv_adapter": null,
+        "inv_adapter_reduction_factor": null,
+        "is_parallel": false,
+        "learn_phm": true,
+        "leave_out": [],
+        "ln_after": false,
+        "ln_before": false,
+        "mh_adapter": false,
+        "non_linearity": "relu",
+        "original_ln_after": true,
+        "original_ln_before": true,
+        "output_adapter": true,
+        "phm_bias": true,
+        "phm_c_init": "normal",
+        "phm_dim": 4,
+        "phm_init_range": 0.0001,
+        "phm_layer": false,
+        "phm_rank": 1,
+        "reduction_factor": 16,
+        "residual_before_ln": true,
+        "scaling": 1.0,
+        "shared_W_phm": false,
+        "shared_phm_rule": true,
+        "use_gating": true
+      }
+    ]
+  },
+  "hidden_size": 512,
+  "model_class": "T5AdapterModel",
+  "model_name": "t5-small",
+  "model_type": "t5",
+  "name": "xsum",
+  "version": "0.1.1"
+}

head_config.json

@@ -0,0 +1,18 @@
+{
+  "config": {
+    "activation_function": null,
+    "bias": false,
+    "head_type": "seq2seq_lm",
+    "label2id": null,
+    "layer_norm": false,
+    "layers": 1,
+    "shift_labels": false,
+    "vocab_size": 32128
+  },
+  "hidden_size": 512,
+  "model_class": "T5AdapterModel",
+  "model_name": "t5-small",
+  "model_type": "t5",
+  "name": "xsum",
+  "version": "0.1.1"
+}

pytorch_adapter.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d9ab432246b2ca00cc5106671b92082bdb31be0a99e6b572de32b1ebee84ad0
+size 44106132

pytorch_model_head.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c62960e655f28561ab8afdea57f4cec119b1ab6a5badf01a1c4bd445716f5d1
+size 65799443