Upload ExtendedMptForCausalLM

config.json

@@ -1,28 +1,14 @@
 {
-  "_name_or_path": "normalcomputing/extended-mind-mpt-30b-chat",
+  "_name_or_path": "mosaicml/mpt-30b-chat",
   "architectures": [
     "ExtendedMptForCausalLM"
   ],
   "attn_config": {
-    "alibi": true,
-    "alibi_bias_max": 8,
-    "attn_impl": "torch",
-    "attn_pdrop": 0,
-    "attn_type": "multihead_attention",
-    "attn_uses_sequence_id": false,
-    "clip_qkv": null,
-    "mask_by_sim": true,
-    "memory_type": "manual",
-    "prefix_lm": false,
-    "qk_ln": false,
-    "sim_threshold": 0.25,
-    "softmax_scale": null,
-    "topk": 10,
-    "use_active_externalism": true
+    "model_type": ""
   },
   "auto_map": {
-    "AutoConfig": "configuration.ExtendedMptConfig",
-    "AutoModelForCausalLM": "modeling.ExtendedMptForCausalLM"
+    "AutoConfig": "mosaicml/mpt-30b-chat--configuration_mpt.MPTConfig",
+    "AutoModelForCausalLM": "mosaicml/mpt-30b-chat--modeling_mpt.MPTForCausalLM"
   },
   "d_model": 7168,
   "emb_pdrop": 0,
@@ -55,7 +41,7 @@
   "tokenizer_name": "sam-mosaic/gpt-neox-20b-chatml",
   "torch_dtype": "float32",
   "transformers_version": "4.33.0",
-  "use_cache": true,
+  "use_cache": false,
   "use_external_mind": true,
   "use_external_mind_by_layer": [
     true,

generation_config.json

@@ -1,6 +1,9 @@
 {
-  "eos_token_id": [0,50278],
-  "pad_token_id":0,
-  "do_sample": false,
+  "_from_model_config": true,
+  "eos_token_id": [
+    0,
+    50278
+  ],
   "transformers_version": "4.33.0",
+  "use_cache": false
 }