Upload SwitchTransformersForConditionalGeneration

config.json

@@ -0,0 +1,74 @@
+{
+  "_name_or_path": "google/switch-base-8",
+  "add_router_probs": false,
+  "architectures": [
+    "SwitchTransformersForConditionalGeneration"
+  ],
+  "batch_prioritized_routing": false,
+  "d_ff": 3072,
+  "d_kv": 64,
+  "d_model": 768,
+  "decoder_sparse_step": 2,
+  "decoder_start_token_id": 0,
+  "dense_act_fn": "relu",
+  "dropout_rate": 0.1,
+  "encoder_sparse_step": 2,
+  "eos_token_id": 1,
+  "expert_capacity": 64,
+  "feed_forward_proj": "relu",
+  "initializer_factor": 1.0,
+  "is_encoder_decoder": true,
+  "is_gated_act": false,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "switch_transformers",
+  "num_decoder_layers": 12,
+  "num_experts": 8,
+  "num_heads": 12,
+  "num_layers": 12,
+  "num_selected_experts": 1,
+  "num_sparse_decoder_layers": 6,
+  "num_sparse_encoder_layers": 6,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "router_aux_loss_coef": 0.001,
+  "router_bias": false,
+  "router_dtype": "float32",
+  "router_ignore_padding_tokens": false,
+  "router_jitter_noise": 0,
+  "router_type": "tokens_masked",
+  "router_z_loss_coef": 0.001,
+  "task_specific_params": {
+    "summarization": {
+      "early_stopping": true,
+      "length_penalty": 2.0,
+      "max_length": 200,
+      "min_length": 30,
+      "no_repeat_ngram_size": 3,
+      "num_beams": 4,
+      "prefix": "summarize: "
+    },
+    "translation_en_to_de": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to German: "
+    },
+    "translation_en_to_fr": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to French: "
+    },
+    "translation_en_to_ro": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to Romanian: "
+    }
+  },
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.27.0.dev0",
+  "use_cache": true,
+  "vocab_size": 32128
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "decoder_start_token_id": 0,
+  "eos_token_id": 1,
+  "pad_token_id": 0,
+  "transformers_version": "4.27.0.dev0"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fe774e6257e4fa97d0560629fd3aa791acf3e7fc27f6b573979be7a1cd4e18d
+size 1238978145