888aa0346f6ad1bfdeb436d8fc737cac86d4c5d19b684432a1b88c8c9fbc61a3

README.md

@@ -34,7 +34,7 @@ tags:
 
 ## Results
 
-Detailed efficiency metrics coming soon!
+![image info](./plots.png)
 
 **Frequently Asked Questions**
 - ***How does the compression work?*** The model is compressed with llm-int8.

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/tmp/tmpspfclkor",
+  "_name_or_path": "/tmp/tmplq5oo90c",
   "architectures": [
     "LlamaForCausalLM"
   ],
@@ -21,7 +21,7 @@
   "quantization_config": {
     "bnb_4bit_compute_dtype": "bfloat16",
     "bnb_4bit_quant_type": "fp4",
-    "bnb_4bit_use_double_quant": true,
+    "bnb_4bit_use_double_quant": false,
     "llm_int8_enable_fp32_cpu_offload": false,
     "llm_int8_has_fp16_weight": false,
     "llm_int8_skip_modules": [

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dab394c1adbca0e0da02bc71f7ed20ff55d1410091d7175f78d5bb0a4618992f
-size 1048758259
+oid sha256:647313ff4fcdc83899aeea8184817f407a2e2ffdfd5fd2cf27016d86cfe7a155
+size 1106034112

smash_config.json

@@ -2,18 +2,18 @@
     "api_key": null,
     "verify_url": "http://johnrachwan.pythonanywhere.com",
     "smash_config": {
-        "pruners": "None",
-        "factorizers": "None",
+        "pruners": "[]",
+        "factorizers": "[]",
         "quantizers": "['llm-int8']",
-        "compilers": "None",
+        "compilers": "[]",
         "task": "text_text_generation",
         "device": "cuda",
-        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsunczu0da",
+        "cache_dir": "/ceph/hdd/staff/charpent/.cache/models7ldbxd23",
         "batch_size": 1,
         "model_name": "42dot/42dot_LLM-PLM-1.3B",
         "pruning_ratio": 0.0,
         "n_quantization_bits": 4,
-        "output_deviation": 0.005,
+        "output_deviation": 0.01,
         "max_batch_size": 1,
         "qtype_weight": "torch.qint8",
         "qtype_activation": "torch.quint8",