677831e230c552374dd251bbaab42afe2d0a8135aab9e4ea04c169de10ce767d

README.md

@@ -31,7 +31,7 @@ tags:
 - Contact us and tell us which model to compress next [here](https://www.pruna.ai/contact).
 - Request access to easily compress your *own* AI models [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
 - Read the documentations to know more [here](https://pruna-ai-pruna.readthedocs-hosted.com/en/latest/)
-- Join Pruna AI community on Discord [here](https://discord.gg/rskEr4BZJx) to share feedback/suggestions or get help.
+- Join Pruna AI community on Discord [here](https://discord.gg/CP4VSgck) to share feedback/suggestions or get help.
 
 ## Results
 
@@ -40,7 +40,7 @@ tags:
 **Frequently Asked Questions**
 - ***How does the compression work?*** The model is compressed with hqq.
 - ***How does the model quality change?*** The quality of the model output might vary compared to the base model.
-- ***How is the model efficiency evaluated?*** These results were obtained on NVIDIA A100-PCIE-40GB with configuration described in `model/smash_config.json` and are obtained after a hardware warmup. The smashed model is directly compared to the original base model. Efficiency results may vary in other settings (e.g. other hardware, image size, batch size, ...). We recommend to directly run them in the use-case conditions to know if the smashed model can benefit you.
+- ***How is the model efficiency evaluated?*** These results were obtained on HARDWARE_NAME with configuration described in `model/smash_config.json` and are obtained after a hardware warmup. The smashed model is directly compared to the original base model. Efficiency results may vary in other settings (e.g. other hardware, image size, batch size, ...). We recommend to directly run them in the use-case conditions to know if the smashed model can benefit you.
 - ***What is the model format?*** We use safetensors.
 - ***What calibration data has been used?*** If needed by the compression method, we used WikiText as the calibration data.
 - ***What is the naming convention for Pruna Huggingface models?*** We take the original model name and append "turbo", "tiny", or "green" if the smashed model has a measured inference speed, inference memory, or inference energy consumption which is less than 90% of the original base model.

config.json

@@ -3,6 +3,7 @@
   "architectures": [
     "Phi3ForCausalLM"
   ],
+  "attention_bias": false,
   "attention_dropout": 0.0,
   "auto_map": {
     "AutoConfig": "microsoft/Phi-3-mini-128k-instruct--configuration_phi3.Phi3Config",
@@ -26,71 +27,64 @@
   "rms_norm_eps": 1e-05,
   "rope_scaling": {
     "long_factor": [
-      1.0299999713897705,
-      1.0499999523162842,
-      1.0499999523162842,
-      1.0799999237060547,
-      1.2299998998641968,
-      1.2299998998641968,
-      1.2999999523162842,
-      1.4499999284744263,
-      1.5999999046325684,
-      1.6499998569488525,
-      1.8999998569488525,
-      2.859999895095825,
-      3.68999981880188,
-      5.419999599456787,
-      5.489999771118164,
-      5.489999771118164,
-      9.09000015258789,
-      11.579999923706055,
-      15.65999984741211,
-      15.769999504089355,
-      15.789999961853027,
-      18.360000610351562,
-      21.989999771118164,
-      23.079999923706055,
-      30.009998321533203,
-      32.35000228881836,
-      32.590003967285156,
-      35.56000518798828,
-      39.95000457763672,
-      53.840003967285156,
-      56.20000457763672,
-      57.95000457763672,
-      59.29000473022461,
-      59.77000427246094,
-      59.920005798339844,
-      61.190006256103516,
-      61.96000671386719,
-      62.50000762939453,
-      63.3700065612793,
-      63.48000717163086,
-      63.48000717163086,
-      63.66000747680664,
-      63.850006103515625,
-      64.08000946044922,
-      64.760009765625,
-      64.80001068115234,
-      64.81001281738281,
-      64.81001281738281
+      1.0700000524520874,
+      1.1200000047683716,
+      1.149999976158142,
+      1.4199999570846558,
+      1.5699999332427979,
+      1.7999999523162842,
+      2.129999876022339,
+      2.129999876022339,
+      3.009999990463257,
+      5.910000324249268,
+      6.950000286102295,
+      9.070000648498535,
+      9.930000305175781,
+      10.710000038146973,
+      11.130000114440918,
+      14.609999656677246,
+      15.409998893737793,
+      19.809999465942383,
+      37.279998779296875,
+      38.279998779296875,
+      38.599998474121094,
+      40.12000274658203,
+      46.20000457763672,
+      50.940006256103516,
+      53.66000747680664,
+      54.9373893737793,
+      56.89738845825195,
+      57.28738784790039,
+      59.98738479614258,
+      60.86738586425781,
+      60.887386322021484,
+      61.71739196777344,
+      62.91739273071289,
+      62.957393646240234,
+      63.41739273071289,
+      63.8173942565918,
+      63.83739471435547,
+      63.897396087646484,
+      63.93739700317383,
+      64.06739807128906,
+      64.11434936523438,
+      64.12435150146484,
+      64.15435028076172,
+      64.19435119628906,
+      64.24435424804688,
+      64.57435607910156,
+      64.69000244140625,
+      64.76000213623047
     ],
     "short_factor": [
-      1.05,
-      1.05,
-      1.05,
       1.1,
       1.1,
-      1.1500000000000001,
-      1.2000000000000002,
-      1.2500000000000002,
+      1.1,
       1.3000000000000003,
       1.3500000000000003,
-      1.5000000000000004,
-      2.000000000000001,
-      2.000000000000001,
-      2.000000000000001,
-      2.000000000000001,
+      1.3500000000000003,
+      1.4000000000000004,
+      1.5500000000000005,
       2.000000000000001,
       2.000000000000001,
       2.000000000000001,
@@ -111,27 +105,34 @@
       2.0500000000000007,
       2.0500000000000007,
       2.0500000000000007,
+      2.0500000000000007,
+      2.0500000000000007,
+      2.0500000000000007,
       2.1000000000000005,
       2.1000000000000005,
-      2.1000000000000005,
-      2.1500000000000004,
       2.1500000000000004,
-      2.3499999999999996,
-      2.549999999999999,
-      2.5999999999999988,
-      2.5999999999999988,
+      2.25,
+      2.25,
+      2.25,
+      2.25,
+      2.25,
+      2.3999999999999995,
+      2.4499999999999993,
+      2.499999999999999,
+      2.6999999999999984,
+      2.6999999999999984,
       2.7499999999999982,
-      2.849999999999998,
-      2.849999999999998,
-      2.9499999999999975
+      2.799999999999998,
+      2.8999999999999977,
+      3.049999999999997
     ],
-    "type": "su"
+    "type": "longrope"
   },
   "rope_theta": 10000.0,
   "sliding_window": 262144,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.40.0",
+  "transformers_version": "4.42.4",
   "use_cache": true,
   "vocab_size": 32064
 }

qmodel.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b593bf5801aa0f5eddc066f1c3b01e4a490acd30ea1ae869d275c352a329d117
+oid sha256:751c4e665579b41d169cbdcc855fb95ebb1e91b75ea453308660262dd5c5e4da
 size 932545573

smash_config.json

@@ -14,7 +14,7 @@
         "controlnet": "None",
         "unet_dim": 4,
         "device": "cuda",
-        "cache_dir": "/ceph/hdd/staff/charpent/.cache/models8pnn8345",
+        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsylk47uod",
         "batch_size": 1,
         "model_name": "microsoft/Phi-3-mini-128k-instruct",
         "task": "text_text_generation",

tokenizer.json

@@ -26,9 +26,9 @@
       "content": "</s>",
       "single_word": false,
       "lstrip": false,
-      "rstrip": false,
+      "rstrip": true,
       "normalized": false,
-      "special": true
+      "special": false
     },
     {
       "id": 32000,
@@ -44,7 +44,7 @@
       "content": "<|assistant|>",
       "single_word": false,
       "lstrip": false,
-      "rstrip": false,
+      "rstrip": true,
       "normalized": false,
       "special": true
     },
@@ -53,7 +53,7 @@
       "content": "<|placeholder1|>",
       "single_word": false,
       "lstrip": false,
-      "rstrip": false,
+      "rstrip": true,
       "normalized": false,
       "special": true
     },
@@ -62,7 +62,7 @@
       "content": "<|placeholder2|>",
       "single_word": false,
       "lstrip": false,
-      "rstrip": false,
+      "rstrip": true,
       "normalized": false,
       "special": true
     },
@@ -71,7 +71,7 @@
       "content": "<|placeholder3|>",
       "single_word": false,
       "lstrip": false,
-      "rstrip": false,
+      "rstrip": true,
       "normalized": false,
       "special": true
     },
@@ -80,7 +80,7 @@
       "content": "<|placeholder4|>",
       "single_word": false,
       "lstrip": false,
-      "rstrip": false,
+      "rstrip": true,
       "normalized": false,
       "special": true
     },
@@ -89,7 +89,7 @@
       "content": "<|system|>",
       "single_word": false,
       "lstrip": false,
-      "rstrip": false,
+      "rstrip": true,
       "normalized": false,
       "special": true
     },
@@ -98,7 +98,7 @@
       "content": "<|end|>",
       "single_word": false,
       "lstrip": false,
-      "rstrip": false,
+      "rstrip": true,
       "normalized": false,
       "special": true
     },
@@ -107,7 +107,7 @@
       "content": "<|placeholder5|>",
       "single_word": false,
       "lstrip": false,
-      "rstrip": false,
+      "rstrip": true,
       "normalized": false,
       "special": true
     },
@@ -116,7 +116,7 @@
       "content": "<|placeholder6|>",
       "single_word": false,
       "lstrip": false,
-      "rstrip": false,
+      "rstrip": true,
       "normalized": false,
       "special": true
     },
@@ -125,7 +125,7 @@
       "content": "<|user|>",
       "single_word": false,
       "lstrip": false,
-      "rstrip": false,
+      "rstrip": true,
       "normalized": false,
       "special": true
     }
@@ -150,12 +150,6 @@
   "post_processor": {
     "type": "TemplateProcessing",
     "single": [
-      {
-        "SpecialToken": {
-          "id": "<s>",
-          "type_id": 0
-        }
-      },
       {
         "Sequence": {
           "id": "A",
@@ -164,24 +158,12 @@
       }
     ],
     "pair": [
-      {
-        "SpecialToken": {
-          "id": "<s>",
-          "type_id": 0
-        }
-      },
       {
         "Sequence": {
           "id": "A",
           "type_id": 0
         }
       },
-      {
-        "SpecialToken": {
-          "id": "<s>",
-          "type_id": 1
-        }
-      },
       {
         "Sequence": {
           "id": "B",
@@ -189,17 +171,7 @@
         }
       }
     ],
-    "special_tokens": {
-      "<s>": {
-        "id": "<s>",
-        "ids": [
-          1
-        ],
-        "tokens": [
-          "<s>"
-        ]
-      }
-    }
+    "special_tokens": {}
   },
   "decoder": {
     "type": "Sequence",

tokenizer_config.json

@@ -1,6 +1,7 @@
 {
-  "add_bos_token": true,
+  "add_bos_token": false,
   "add_eos_token": false,
+  "add_prefix_space": null,
   "added_tokens_decoder": {
     "0": {
       "content": "<unk>",
@@ -22,9 +23,9 @@
       "content": "</s>",
       "lstrip": false,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
-      "special": true
+      "special": false
     },
     "32000": {
       "content": "<|endoftext|>",
@@ -38,7 +39,7 @@
       "content": "<|assistant|>",
       "lstrip": false,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
@@ -46,7 +47,7 @@
       "content": "<|placeholder1|>",
       "lstrip": false,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
@@ -54,7 +55,7 @@
       "content": "<|placeholder2|>",
       "lstrip": false,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
@@ -62,7 +63,7 @@
       "content": "<|placeholder3|>",
       "lstrip": false,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
@@ -70,7 +71,7 @@
       "content": "<|placeholder4|>",
       "lstrip": false,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
@@ -78,7 +79,7 @@
       "content": "<|system|>",
       "lstrip": false,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
@@ -86,7 +87,7 @@
       "content": "<|end|>",
       "lstrip": false,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
@@ -94,7 +95,7 @@
       "content": "<|placeholder5|>",
       "lstrip": false,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
@@ -102,7 +103,7 @@
       "content": "<|placeholder6|>",
       "lstrip": false,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
@@ -110,13 +111,13 @@
       "content": "<|user|>",
       "lstrip": false,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     }
   },
   "bos_token": "<s>",
-  "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] in ['user', 'system']) %}{{'<|user|>' + '\n' + message['content'] + '<|end|>' + '\n' + '<|assistant|>' + '\n'}}{% elif message['role'] == 'assistant' %}{{message['content'] + '<|end|>' + '\n'}}{% endif %}{% endfor %}",
+  "chat_template": "{% for message in messages %}{% if message['role'] == 'system' %}{{'<|system|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'user' %}{{'<|user|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'assistant' %}{{'<|assistant|>\n' + message['content'] + '<|end|>\n'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>\n' }}{% else %}{{ eos_token }}{% endif %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|endoftext|>",
   "legacy": false,