Upload 45 files

.gitattributes

@@ -34,3 +34,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 *.axmodel filter=lfs diff=lfs merge=lfs -text
+main_axcl_aarch64 filter=lfs diff=lfs merge=lfs -text
+main_axcl_x86 filter=lfs diff=lfs merge=lfs -text
+main_prefill filter=lfs diff=lfs merge=lfs -text

main_axcl_aarch64

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb111fc00c54abb6142a8f44df087bf104c8150a1cefa6be55c6b174b932c4ec
+size 999008

main_axcl_x86

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cba0be8df523e351789fcfa745772423096f52a3b0a760f8d8c9f5b8bb2ec82
+size 1022384

main_prefill

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90026f1e071aef781ccc9bc313db3805749b3e234bf03d578c9f97ae2e60abd2
+size 953544

post_config.json

@@ -0,0 +1,14 @@
+{
+    "enable_temperature" : true,
+    "temperature" : 0.9,
+    
+    "enable_repetition_penalty" : false,
+    "repetition_penalty" : 1.2,
+    "penalty_window" : 20,
+
+    "enable_top_p_sampling" : false,
+    "top_p" : 0.8,
+
+    "enable_top_k_sampling" : true,
+    "top_k" : 10
+}

run_smollm2_360m_ax650.sh

@@ -0,0 +1,14 @@
+./main_prefill \
+--template_filename_axmodel "smollm2-360m-ax650/llama_p128_l%d_together.axmodel" \
+--axmodel_num 32 \
+--tokenizer_type 2 \
+--filename_tokenizer_model "http://127.0.0.1:12345" \
+--bos 0 --eos 0 \
+--filename_post_axmodel "smollm2-360m-ax650/llama_post.axmodel" \
+--filename_tokens_embed "smollm2-360m-ax650/model.embed_tokens.weight.bfloat16.bin" \
+--tokens_embed_num 49152 \
+--tokens_embed_size 960 \
+--use_mmap_load_embed 1 \
+--live_print 1 \
+--continue 1 \
+--prompt "$1"

run_smollm2_360m_axcl_aarch64.sh

@@ -0,0 +1,14 @@
+./main_axcl_aarch64 \
+--template_filename_axmodel "smollm2-360m-ax650/llama_p128_l%d_together.axmodel" \
+--axmodel_num 32 \
+--tokenizer_type 2 \
+--filename_tokenizer_model "http://127.0.0.1:12345" \
+--bos 0 --eos 0 \
+--filename_post_axmodel "smollm2-360m-ax650/llama_post.axmodel" \
+--filename_tokens_embed "smollm2-360m-ax650/model.embed_tokens.weight.bfloat16.bin" \
+--tokens_embed_num 49152 \
+--tokens_embed_size 960 \
+--use_mmap_load_embed 0 \
+--live_print 1 \
+--continue 1 \
+--prompt "$1"

run_smollm2_360m_axcl_x86.sh

@@ -0,0 +1,14 @@
+./main_axcl_x86 \
+--template_filename_axmodel "smollm2-360m-ax650/llama_p128_l%d_together.axmodel" \
+--axmodel_num 32 \
+--tokenizer_type 2 \
+--filename_tokenizer_model "http://127.0.0.1:12345" \
+--bos 0 --eos 0 \
+--filename_post_axmodel "smollm2-360m-ax650/llama_post.axmodel" \
+--filename_tokens_embed "smollm2-360m-ax650/model.embed_tokens.weight.bfloat16.bin" \
+--tokens_embed_num 49152 \
+--tokens_embed_size 960 \
+--use_mmap_load_embed 0 \
+--live_print 1 \
+--continue 1 \
+--prompt "$1"

smollm2-360m-ax650/llama_p128_l0_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:489febc320a376e049b6fa6f059dcf423174513da6544b89b36e89cbf8900fd7
+size 11218690

smollm2-360m-ax650/llama_p128_l10_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:676db9c8def427b8d24e89154d7c54623ad96d550a74b93392af5b2351609dc2
+size 11218690

smollm2-360m-ax650/llama_p128_l11_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a572f7f626eb436047c9fdcc14e12f448faebaa7ac7ec54710e25f8283a2ae91
+size 11218690

smollm2-360m-ax650/llama_p128_l12_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19aac1ab81b32be3d2b6c32bb30b59157d48e85ece6c2aac36099ea371082b35
+size 11218690

smollm2-360m-ax650/llama_p128_l13_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d1ae96d861cafeed478b4953b0a17f018bf2d3d7b65704509ec4462f0d931ba
+size 11218690

smollm2-360m-ax650/llama_p128_l14_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aaeefd23274a3aab6018781683b03c097fd07d565c779b9c01cf65e30d42801f
+size 11218690

smollm2-360m-ax650/llama_p128_l15_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84bd88ec4ce38d35ca28996c1b178953b3ea2d97eb4a1a6941ed2a4a61b678fc
+size 11218690

smollm2-360m-ax650/llama_p128_l16_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf68c070d51d404f1c1c861e8e5d9e7e7c3421ee59a7790f30050a056933c62e
+size 11218690

smollm2-360m-ax650/llama_p128_l17_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90dabb01513932459c2c344f8aee8f8c53d6869b6a5ab91251ef18fd139a619c
+size 11218690

smollm2-360m-ax650/llama_p128_l18_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5ed70f209cfb6859971bfc6acc289e6768fd160799a54a4a30e0bcf7860bb05
+size 11218690

smollm2-360m-ax650/llama_p128_l19_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13d65f6805ed97aea4c813b7bcf90821ec0ad6ce7e0a1be0d385554cb8211c34
+size 11218690

smollm2-360m-ax650/llama_p128_l1_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:995f165341d9808fa11f754236041555edf933571aec9d9938bf480ac8f05328
+size 11218690

smollm2-360m-ax650/llama_p128_l20_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7da1158e22b5ca6a3332949d1c456a62574cd0ce274b0c247b685856f3329f31
+size 11218690

smollm2-360m-ax650/llama_p128_l21_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f2ae999d342e4a76292a0117f9604187c8985cd07cb2ef11e4f57c7381bcc0a
+size 11218690

smollm2-360m-ax650/llama_p128_l22_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1b428789458de91e3fb4cd95d6e58c400e06795dbcc3cffea1e78c0d58d902a
+size 11218690

smollm2-360m-ax650/llama_p128_l23_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad53076c95797aebf892595a2b7d88fb3e712e27efdfd92a54561c750a88124d
+size 11218690

smollm2-360m-ax650/llama_p128_l24_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0bcf51f105a596b88a8d06f8095cad551b337a4384afb645fb554d0e346a7382
+size 11218690

smollm2-360m-ax650/llama_p128_l25_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:002e44afa0542f838bf9fde6d217ac92c6855c5bfc9c8802e8e80c5fb0509687
+size 11218690

smollm2-360m-ax650/llama_p128_l26_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7aa19ca187a16f2f08afc80ff9f755e78cf18868f45b699c065c2bb1fb194290
+size 11218690

smollm2-360m-ax650/llama_p128_l27_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da4bf6b5fdc6abd69f7cc6bd259df7685972de282f07e66def6236a209623820
+size 11218690

smollm2-360m-ax650/llama_p128_l28_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:416d24cf85eb4eab1cb5bc19cea300318b4a0d34b63ccf591f5264ef827e9bfa
+size 11218690

smollm2-360m-ax650/llama_p128_l29_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:579e1b979e97247aed78dabe564d50b5bdbc81c0edada4046c81c43cd76b645e
+size 11218690

smollm2-360m-ax650/llama_p128_l2_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fabbd458282bb1dc52006a1c96c68fcdcb7c8b686a557a1da08dc15fc29e4b80
+size 11218690

smollm2-360m-ax650/llama_p128_l30_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7bd3399a0a851f05116de727c0a28838dace9f8096876fe782cf8383fe314a5
+size 11218690

smollm2-360m-ax650/llama_p128_l31_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5078029f054ba1c4a2ecf0dd0919865084017b39d8c790049227bdffb1d048c
+size 11218690

smollm2-360m-ax650/llama_p128_l3_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0983d31036313e367a7cf91bb771bee3cb1854e07baf4068bff123ffd296c98e
+size 11218690

smollm2-360m-ax650/llama_p128_l4_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14b9402ae0d8f05d08b525fc571f641f219138e92f2ce9b82ac89cb6ffc15910
+size 11218690

smollm2-360m-ax650/llama_p128_l5_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d2143709587c18dce172eb8297fa7ff57bdf397a1a0c4f3e9fd56d63720dc2a
+size 11218690

smollm2-360m-ax650/llama_p128_l6_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff12125e92ed34c5bfe5efd7232229d69b23956248236e34d3357c711a78736a
+size 11218690

smollm2-360m-ax650/llama_p128_l7_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eb6216f8f4531364348e7e85403e305103eef19ac7ba18b33b7fff0afdc2d9e
+size 11218690

smollm2-360m-ax650/llama_p128_l8_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af8cebf4cf6698ebfdb07663a18cc9bc046ae52944f51902315a3a7df72c318c
+size 11218690

smollm2-360m-ax650/llama_p128_l9_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33b7bbe08d59ee7d2c187edf8c85cee8aee464fe11fb1c5608a44369fb83907c
+size 11218690

smollm2-360m-ax650/llama_post.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f891122c5016004037af8b7dd70edd218729548301e432425d0e9c2305610adb
+size 51443158

smollm2-360m-ax650/model.embed_tokens.weight.bfloat16.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc8a09d622a836b0b05ee1d8144054f8ac8b2d7e239e6957aae3b52134c12cb3
+size 94371840

smollm2_tokenizer.py

@@ -0,0 +1,133 @@
+from transformers import AutoTokenizer, PreTrainedTokenizerFast
+from http.server import HTTPServer, BaseHTTPRequestHandler
+import json
+import argparse
+
+
+class Tokenizer_Http():
+
+    def __init__(self):
+        model_id = "smollm2_tokenizer"
+        self.tokenizer = AutoTokenizer.from_pretrained(model_id)
+
+    def encode(self, prompt):
+        messages = [
+            #{"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."},
+            {"role": "user", "content": prompt}
+        ]
+        text = self.tokenizer.apply_chat_template(
+            messages,
+            tokenize=False,
+            add_generation_prompt=True
+        )
+        print(text)
+        token_ids = self.tokenizer.encode(text)
+        return token_ids
+
+    def decode(self, token_ids):
+        return self.tokenizer.decode(token_ids)
+
+    @property
+    def bos_id(self):
+        return self.tokenizer.bos_token_id
+
+    @property
+    def eos_id(self):
+        return self.tokenizer.eos_token_id
+    
+    @property
+    def bos_token(self):
+        return self.tokenizer.bos_token
+
+    @property
+    def eos_token(self):
+        return self.tokenizer.eos_token
+
+
+tokenizer = Tokenizer_Http()
+
+print(tokenizer.bos_id, tokenizer.bos_token, tokenizer.eos_id, tokenizer.eos_token)
+print(tokenizer.encode("hello world"))
+
+
+class Request(BaseHTTPRequestHandler):
+    #通过类继承，新定义类
+    timeout = 5
+    server_version = 'Apache'
+
+    def do_GET(self):
+        print(self.path)
+        #在新类中定义get的内容（当客户端向该服务端使用get请求时，本服务端将如下运行）
+        self.send_response(200)
+        self.send_header("type", "get")  #设置响应头，可省略或设置多个
+        self.end_headers()
+
+        if self.path == '/bos_id':
+            bos_id = tokenizer.bos_id
+            # print(bos_id)
+            # to json
+            if bos_id is None:
+                msg = json.dumps({'bos_id': -1})
+            else:
+                msg = json.dumps({'bos_id': bos_id})
+        elif self.path == '/eos_id':
+            eos_id = tokenizer.eos_id
+            if eos_id is None:
+                msg = json.dumps({'eos_id': -1})
+            else:
+                msg = json.dumps({'eos_id': eos_id})
+        else:
+            msg = 'error'
+
+        print(msg)
+        msg = str(msg).encode()  #转为str再转为byte格式
+
+        self.wfile.write(msg)  #将byte格式的信息返回给客户端
+
+    def do_POST(self):
+        #在新类中定义post的内容（当客户端向该服务端使用post请求时，本服务端将如下运行）
+        data = self.rfile.read(int(
+            self.headers['content-length']))  #获取从客户端传入的参数（byte格式）
+        data = data.decode()  #将byte格式转为str格式
+
+        self.send_response(200)
+        self.send_header("type", "post")  #设置响应头，可省略或设置多个
+        self.end_headers()
+
+        if self.path == '/encode':
+            req = json.loads(data)
+            prompt = req['text']
+
+            token_ids = tokenizer.encode(prompt)
+            if token_ids is None:
+                msg = json.dumps({'token_ids': -1})
+            else:
+                msg = json.dumps({'token_ids': token_ids})
+
+        elif self.path == '/decode':
+            req = json.loads(data)
+            token_ids = req['token_ids']
+            text = tokenizer.decode(token_ids)
+            if text is None:
+                msg = json.dumps({'text': ""})
+            else:
+                msg = json.dumps({'text': text})
+        else:
+            msg = 'error'
+        print(msg)
+        msg = str(msg).encode()  #转为str再转为byte格式
+
+        self.wfile.write(msg)  #将byte格式的信息返回给客户端
+
+
+if __name__ == "__main__":
+
+    args = argparse.ArgumentParser()
+    args.add_argument('--host', type=str, default='localhost')
+    args.add_argument('--port', type=int, default=8080)
+    args = args.parse_args()
+
+    host = (args.host, args.port)  #设定地址与端口号，'localhost'等价于'127.0.0.1'
+    print('http://%s:%s' % host)
+    server = HTTPServer(host, Request)  #根据地址端口号和新定义的类，创建服务器实例
+    server.serve_forever()  #开启服务

smollm2_tokenizer/tokenizer_config.json

@@ -0,0 +1,154 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<repo_name>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<reponame>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<file_sep>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "<filename>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "<gh_stars>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "<issue_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "<issue_comment>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<issue_closed>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "<jupyter_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "<jupyter_text>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "<jupyter_code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<jupyter_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<jupyter_script>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<empty_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": "<|im_start|>",
+  "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful AI assistant named SmolLM, trained by Hugging Face<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "model_max_length": 8192,
+  "pad_token": "<|im_end|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>",
+  "vocab_size": 49152
+}