Upload HGRNForCausalLM

README.md

@@ -8,4 +8,4 @@ license: mit
 datasets:
 - cerebras/SlimPajama-627B
 library_name: fla
----
+---

config.json

@@ -3,6 +3,7 @@
   "architectures": [
     "HGRNForCausalLM"
   ],
+  "attn": null,
   "attn_mode": "fused_recurrent",
   "bos_token_id": 1,
   "conv_size": 4,
@@ -10,6 +11,7 @@
   "eos_token_id": 2,
   "expand_ratio": 1,
   "fuse_cross_entropy": true,
+  "fuse_norm": true,
   "hidden_act": "swish",
   "hidden_ratio": 4,
   "hidden_size": 2048,
@@ -24,7 +26,7 @@
   "share_conv_kernel": true,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.45.2",
+  "transformers_version": "4.48.2",
   "use_cache": true,
   "use_lower_bound": true,
   "use_short_conv": false,

generation_config.json

@@ -2,5 +2,6 @@
   "_from_model_config": true,
   "bos_token_id": 1,
   "eos_token_id": 2,
-  "transformers_version": "4.45.2"
+  "pad_token_id": 2,
+  "transformers_version": "4.48.2"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c242dd37b51c8b16cfdb8fa7b239bf41e4a172bc93236d3c0997c5fc7b1fc0fb
-size 2728816304
+oid sha256:480ed12b06d2ebfc604210c90ecb66e922d5825e7260ad49cd1061ef7d52a987
+size 2728818968