Updating pruned70+quantized model

README.md

@@ -50,8 +50,8 @@ Model evaluation metrics and results.
 
 | Benchmark                                      | Metric        | Llama-2-7b-evolcodealpaca  | Llama-2-7b-pruned70-retrained-evolcodealpaca-quant-ds |
 |------------------------------------------------|---------------|-------------|-------------------------------|
-| [HumanEval](https://arxiv.org/abs/2107.03374)  | pass@1        | 32.03       | 34.76                         |
+| [HumanEval](https://arxiv.org/abs/2107.03374)  | pass@1        | 32.03       | 35.02                         |
 
 ## Help
 
-For further support, and discussions on these models and AI in general, join [Neural Magic's Slack Community](https://join.slack.com/t/discuss-neuralmagic/shared_invite/zt-q1a1cnvo-YBoICSIw3L1dmQpjBeDurQ)
+For further support, and discussions on these models and AI in general, join [Neural Magic's Slack Community](https://join.slack.com/t/discuss-neuralmagic/shared_invite/zt-q1a1cnvo-YBoICSIw3L1dmQpjBeDurQ)

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/home/abhinav/src/llama-recipes/llama_7b_evol_codealpaca_sparse/sparse-sft_cerebras_70sp_uniform_llama_lr5e-4_epochs1_gradclipFalse_cosine_nodistillation-/combined/",
+  "_name_or_path": "neuralmagic/Llama-2-7b-pruned70-retrained-evolcodealpaca-quant-ds",
   "architectures": [
     "LlamaForCausalLM"
   ],
@@ -21,8 +21,9 @@
   "rope_scaling": null,
   "rope_theta": 10000.0,
   "tie_word_embeddings": false,
-  "torch_dtype": "float32",
-  "transformers_version": "1.7.0.20240313",
+  "tokenizer_class": "LlamaTokenizerFast",
+  "torch_dtype": "float16",
+  "transformers_version": "1.7.0.20240504",
   "use_cache": true,
   "vocab_size": 32000
 }

model-orig.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80041fb9222105cf527dd41b05dd62bfea5d0a4632c3ecc11d51bd688dee2525
-size 1222296
+oid sha256:6e88bf46725811cb455886e4fa944eb5c309512b8793bc4d219e505ab5aaa18b
+size 1222611

model.data

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ea54ad705a5d602bedd74eb1bf6855f0f109c1b275b149361ce8a0d126630533
+oid sha256:365753b3f8cd015b79c51aca5caa0e56d208f731a5f4599720ee6f9be3c6f82b
 size 7154772992

model.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3d208294086849085cced1db1ae81ba4720f449452e1c0eadfc8ec6234170df3
-size 1207241
+oid sha256:d0583ad67585c0fd1758a3d3845814bf5158a5cdbb928b9999f98177b7cf6534
+size 1207556

special_tokens_map.json

@@ -13,13 +13,7 @@
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": {
-    "content": "</s>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  },
+  "pad_token": "</s>",
   "unk_token": {
     "content": "<unk>",
     "lstrip": false,

tokenizer.json

@@ -2,7 +2,7 @@
   "version": "1.0",
   "truncation": {
     "direction": "Right",
-    "max_length": 2048,
+    "max_length": 384,
     "strategy": "LongestFirst",
     "stride": 0
   },

tokenizer_config.json

@@ -29,15 +29,11 @@
   "clean_up_tokenization_spaces": false,
   "eos_token": "</s>",
   "legacy": false,
-  "max_length": 2048,
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "</s>",
   "padding_side": "right",
   "sp_model_kwargs": {},
-  "stride": 0,
   "tokenizer_class": "LlamaTokenizer",
-  "truncation_side": "right",
-  "truncation_strategy": "longest_first",
   "unk_token": "<unk>",
   "use_default_system_prompt": false
 }