Upload folder using huggingface_hub

config.json

@@ -3,22 +3,17 @@
     "LlamaForCausalLM"
   ],
   "attention_bias": false,
-  "attention_dropout": 0.0,
   "bos_token_id": 1,
-  "dtype": "bfloat16",
   "eos_token_id": 2,
-  "head_dim": 64,
   "hidden_act": "silu",
   "hidden_size": 2048,
   "initializer_range": 0.02,
   "intermediate_size": 5632,
   "max_position_embeddings": 2048,
-  "mlp_bias": false,
   "model_type": "llama",
   "num_attention_heads": 32,
   "num_hidden_layers": 22,
   "num_key_value_heads": 4,
-  "pad_token_id": null,
   "pretraining_tp": 1,
   "quantization_config": {
     "config_groups": {
@@ -30,53 +25,37 @@
           "re:.*self_attn\\.(k|q|o|v)_proj$"
         ],
         "weights": {
-          "actorder": null,
-          "block_structure": null,
           "dynamic": false,
           "group_size": 128,
           "num_bits": 4,
           "observer": "memoryless_minmax",
-          "observer_kwargs": {},
-          "scale_dtype": null,
           "strategy": "group",
           "symmetric": true,
-          "type": "int",
-          "zp_dtype": null
+          "type": "int"
         }
       },
       "group_1": {
         "format": "int-quantized",
         "input_activations": {
-          "actorder": null,
-          "block_structure": null,
           "dynamic": true,
           "group_size": null,
           "num_bits": 8,
-          "observer": null,
-          "observer_kwargs": {},
-          "scale_dtype": null,
           "strategy": "token",
           "symmetric": true,
-          "type": "int",
-          "zp_dtype": null
+          "type": "int"
         },
         "output_activations": null,
         "targets": [
           "re:.*mlp\\.(down|gate|up)_proj$"
         ],
         "weights": {
-          "actorder": null,
-          "block_structure": null,
           "dynamic": false,
           "group_size": null,
           "num_bits": 8,
           "observer": "memoryless_minmax",
-          "observer_kwargs": {},
-          "scale_dtype": null,
           "strategy": "channel",
           "symmetric": true,
-          "type": "int",
-          "zp_dtype": null
+          "type": "int"
         }
       }
     },
@@ -86,19 +65,19 @@
       "lm_head"
     ],
     "kv_cache_scheme": null,
+    "moe_sparsity_config": null,
     "quant_method": "compressed-tensors",
     "quantization_status": "compressed",
     "sparsity_config": {},
     "transform_config": {},
-    "version": "0.17.2.dev0"
+    "version": "0.20.0"
   },
   "rms_norm_eps": 1e-05,
-  "rope_parameters": {
-    "rope_theta": 10000.0,
-    "rope_type": "default"
-  },
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
   "tie_word_embeddings": false,
-  "transformers_version": "5.10.1",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.35.0",
   "use_cache": true,
   "vocab_size": 32000
 }

generation_config.json

@@ -3,5 +3,5 @@
   "eos_token_id": 2,
   "max_length": 2048,
   "pad_token_id": 0,
-  "transformers_version": "5.10.1"
+  "transformers_version": "5.21.0"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87434fe357dde7cfb5b43077d25b85ccd492bdd6388c7d8958e1e48de38564b0
+oid sha256:8b201ca0c98be2f789f5ac0bbd6bfb7df634b4bd8b038e3f6f1051f86fae6876
 size 1131284624

recipe.yaml

@@ -1,6 +1,7 @@
 quant_stage:
   quant_modifiers:
     AWQModifier:
+      requires_calibration_data: true
       mappings:
       - smooth_layer: re:.*input_layernorm$
         balance_layers: ['re:.*q_proj$', 're:.*k_proj$', 're:.*v_proj$']
@@ -12,11 +13,13 @@ quant_stage:
       ignore: [lm_head]
       scheme: W4A16
       bypass_divisibility_checks: false
+      requires_calibration_data: false
     GPTQModifier:
       targets: ['re:.*mlp\.(down|gate|up)_proj$']
       ignore: [lm_head]
       scheme: W8A8
       bypass_divisibility_checks: false
+      requires_calibration_data: true
       block_size: 128
       dampening_frac: 0.001
-      offload_hessians: false
+      batched_quantization: auto