Upload folder using huggingface_hub

config.json

@@ -0,0 +1,30 @@
+{
+  "architectures": [
+    "NanoChatForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "dtype": "bfloat16",
+  "eos_token_id": 1,
+  "final_logit_softcapping": 15.0,
+  "hidden_act": "relu2",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 2048,
+  "model_type": "nanochat",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 16,
+  "pad_token_id": 1,
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 10000.0,
+    "rope_type": "default"
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.0.0",
+  "use_cache": true,
+  "vocab_size": 65536
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 1,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "pad_token_id": 1,
+  "transformers_version": "5.0.0",
+  "use_cache": true
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1dbeeefcf73a06954ddaa7a1eb980b352f8c190d5876179fd3c496ff7ec46d7c
+size 3758118584

tiktoken/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:226e3db7786a5f498f9bc99b28a1df59298d70f4b21adc30f29e565e8dae03c5
+size 1139495