Upload 5 files

.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

config.json

@@ -0,0 +1,63 @@
+{
+  "_name_or_path": "pico-7b",
+  "architectures": [
+    "PiCoForCausalLM"
+  ],
+  "model_type": "pico",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.0",
+  "auto_map": {
+    "AutoConfig": "pico.configuration.PiCoConfig",
+    "AutoModelForCausalLM": "pico.modeling.PiCoForCausalLM"
+  },
+
+  "vocab_size": 152064,
+  "hidden_size": 2048,
+  "intermediate_size": 5888,
+  "num_hidden_layers": 30,
+  "num_attention_heads": 16,
+  "num_key_value_heads": 4,
+  "head_dim": 128,
+
+  "hidden_act": "silu",
+  "max_position_embeddings": 131072,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 10000.0,
+  "rope_scaling": null,
+  "tie_word_embeddings": false,
+  "use_cache": false,
+  "sliding_window": 4096,
+  "attention_dropout": 0.0,
+  "attention_bias": false,
+  "mlp_bias": false,
+  "initializer_range": 0.02,
+  "pretraining_tp": 1,
+
+  "num_experts": 8,
+  "num_experts_per_tok": 2,
+  "num_shared_experts": 1,
+  "moe_layers": [0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28],
+  "router_aux_loss_coef": 0.001,
+  "router_jitter_noise": 0.01,
+
+  "use_hierarchical_routing": true,
+  "num_routing_levels": 2,
+  "expert_capacity_factor": 1.25,
+
+  "use_expert_compression": true,
+  "expert_compression_ratio": 0.5,
+
+  "use_meta_router": true,
+  "meta_router_hidden_size": 256,
+  "meta_router_learning_rate": 0.01,
+
+  "use_shared_knowledge": true,
+  "shared_knowledge_size": 2048,
+
+  "use_dynamic_expert_pruning": true,
+  "expert_pruning_threshold": 0.01,
+
+  "pad_token_id": 0,
+  "bos_token_id": 1,
+  "eos_token_id": 2
+}

generation_config.json

@@ -0,0 +1,47 @@
+{
+  "_from_model_config": true,
+  "model_type": "pico",
+  "transformers_version": "4.46.0",
+
+  "do_sample": true,
+  "temperature": 0.7,
+  "top_k": 50,
+  "top_p": 0.9,
+  "max_new_tokens": 512,
+
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "pad_token_id": 0,
+
+  "repetition_penalty": 1.0,
+  "length_penalty": 1.0,
+  "no_repeat_ngram_size": 0,
+  "min_new_tokens": 0,
+  "num_beams": 1,
+  "num_return_sequences": 1,
+  "use_cache": true,
+  "renormalize_logits": true,
+
+  "early_stopping": false,
+  "diversity_penalty": 0.0,
+  "num_beam_groups": 1,
+  "encoder_no_repeat_ngram_size": 0,
+  "bad_words_ids": null,
+  "begin_suppress_tokens": null,
+  "forced_bos_token_id": null,
+  "forced_eos_token_id": null,
+  "suppress_tokens": null,
+  "max_time": null,
+  "guidance_scale": null,
+  "sequence_bias": null,
+
+  "assistant_token_id": 14,
+  "user_token_id": 13,
+  "system_token_id": 15,
+
+  "chat_template_tokens": {
+    "user": "<|user|>",
+    "assistant": "<|assistant|>",
+    "system": "<|system|>"
+  }
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a2b1a42439ba39627390295c25a2e9877b02f1452f5578900de922e369ec856
+size 13891368756

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c7647eb1f65f1b1be1a6cc553a466b2ff0b87d66a2d6394394c2bfb3ed14ce2
+size 11042986

tokenizer_config.json

@@ -0,0 +1,45 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<|bos|>",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "<|cls|>",
+  "eos_token": "<|eos|>",
+  "extra_special_tokens": [
+    "<|think|>",
+    "<|reason|>",
+    "<|summary|>",
+    "<|translate|>",
+    "<|code|>",
+    "<|math|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<|system|>"
+  ],
+  "is_local": true,
+  "local_files_only": false,
+  "mask_token": "<|mask|>",
+  "model_max_length": 131072,
+  "pad_token": "<|pad|>",
+  "sep_token": "<|sep|>",
+  "special_tokens": {
+    "assistant": "<|assistant|>",
+    "bos": "<|bos|>",
+    "cls": "<|cls|>",
+    "eos": "<|eos|>",
+    "mask": "<|mask|>",
+    "pad": "<|pad|>",
+    "sep": "<|sep|>",
+    "system": "<|system|>",
+    "unk": "<|unk|>",
+    "user": "<|user|>",
+    "代码": "<|代码|>",
+    "思考": "<|思考|>",
+    "总结": "<|总结|>",
+    "推理": "<|推理|>",
+    "数学": "<|数学|>",
+    "翻译": "<|翻译|>"
+  },
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "<|unk|>",
+  "vocab_size": 152064
+}