Upload MistralForCausalLM

README.md

@@ -1,7 +1,7 @@
 ---
-license: apache-2.0
-language:
-- en
 datasets:
 - Locutusque/UltraTextbooks
----
+language:
+- en
+license: apache-2.0
+---

config.json

@@ -1,11 +1,12 @@
 {
-  "_name_or_path": "Locutusque/TinyMistral-248M-v2.5",
+  "_name_or_path": "Locutusque/TinyMistral-248M",
   "architectures": [
     "MistralForCausalLM"
   ],
   "attention_dropout": 0.0,
   "bos_token_id": 1,
   "eos_token_id": 2,
+  "head_dim": 32,
   "hidden_act": "silu",
   "hidden_size": 1024,
   "initializer_range": 0.02,
@@ -19,8 +20,8 @@
   "rope_theta": 10000.0,
   "sliding_window": 32,
   "tie_word_embeddings": false,
-  "torch_dtype": "float32",
-  "transformers_version": "4.36.2",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.2",
   "use_cache": true,
   "vocab_size": 32005
 }

generation_config.json

@@ -2,5 +2,5 @@
   "_from_model_config": true,
   "bos_token_id": 1,
   "eos_token_id": 2,
-  "transformers_version": "4.36.2"
+  "transformers_version": "4.45.2"
 }

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb9969350acb49480495906a9dd0a4c7e6ee93eb23f2bd6ca463f61dd87a60ee
+size 496060688