Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

chat_template.jinja

@@ -0,0 +1,12 @@
+{#- mockingbird chat template: Seed-native convention, NO THINKING.
+    Structurally the Instruct template minus <seed:think>/budget machinery.
+    Used both for training (Axolotl chat_template_jinja) and shipping
+    (embedded in tokenizer_config.json so MLX/GGUF conversion carries it). -#}
+{%- for message in messages -%}
+{{- '<seed:bos>' + message['role'] + '\n' -}}
+{{- message['content'] | trim -}}
+{{- '<seed:eos>' -}}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+{{- '<seed:bos>assistant\n' -}}
+{%- endif -%}

config.json

@@ -0,0 +1,43 @@
+{
+    "architectures": [
+        "SeedOssForCausalLM"
+    ],
+    "attention_bias": true,
+    "attention_dropout": 0.1,
+    "attention_out_bias": false,
+    "bos_token_id": 0,
+    "eos_token_id": 2,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 5120,
+    "initializer_range": 0.02,
+    "intermediate_size": 27648,
+    "max_position_embeddings": 524288,
+    "mlp_bias": false,
+    "model_type": "seed_oss",
+    "num_attention_heads": 80,
+    "num_hidden_layers": 64,
+    "num_key_value_heads": 8,
+    "pad_token_id": 1,
+    "quantization": {
+        "group_size": 64,
+        "bits": 3,
+        "mode": "affine"
+    },
+    "quantization_config": {
+        "group_size": 64,
+        "bits": 3,
+        "mode": "affine"
+    },
+    "residual_dropout": 0.1,
+    "rms_norm_eps": 1e-06,
+    "rope_scaling": {
+        "rope_type": "default"
+    },
+    "rope_theta": 10000000.0,
+    "tie_word_embeddings": false,
+    "torch_dtype": "bfloat16",
+    "transformers_version": "4.55.0",
+    "use_cache": true,
+    "vocab_size": 155136
+}

generation_config.json

@@ -0,0 +1,9 @@
+{
+  "bos_token_id": 0,
+  "pad_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.55.0",
+  "do_sample": true,
+  "temperature": 1.0,
+  "top_p": 0.9
+}

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b70a5f01324d1a38be45120cd79726bbd148d5a07df77fc90c4d0411d1fbd8f
+size 5360405275

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5bb1f77fcfe756a7e6c254898424310b86acec5d48fee79482c37b3f4e8c4b54
+size 5322580588

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53f47c929b27431d199320160ee81da18b02f87eb8cd698da89590b74b7029ae
+size 5135574107

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6bd848f52451824a3033a9f1e67eea5b399a13c90f845a332d3a29537e05827
+size 11883696

tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<seed:bos>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<seed:eos>",
+  "is_local": true,
+  "local_files_only": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<seed:pad>",
+  "tokenizer_class": "TokenizersBackend",
+  "chat_template": "{#- mockingbird chat template: Seed-native convention, NO THINKING.\n    Structurally the Instruct template minus <seed:think>/budget machinery.\n    Used both for training (Axolotl chat_template_jinja) and shipping\n    (embedded in tokenizer_config.json so MLX/GGUF conversion carries it). -#}\n{%- for message in messages -%}\n{{- '<seed:bos>' + message['role'] + '\\n' -}}\n{{- message['content'] | trim -}}\n{{- '<seed:eos>' -}}\n{%- endfor -%}\n{%- if add_generation_prompt -%}\n{{- '<seed:bos>assistant\\n' -}}\n{%- endif -%}\n"
+}