CallComply
/

Starling-LM-11B-alpha

Text Generation

Inference Endpoints

text-generation-inference

Model card Files Files and versions Community

perlthoughts commited on Jan 14

Commit

dcb0297

•

1 Parent(s): a746dce

Update config.json

Files changed (1) hide show

config.json +5 -4

config.json CHANGED Viewed

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/content/new-merged",
   "architectures": [
     "MistralForCausalLM"
   ],
@@ -9,14 +9,15 @@
   "hidden_size": 4096,
   "initializer_range": 0.02,
   "intermediate_size": 14336,
-  "max_position_embeddings": 8192,
   "model_type": "mistral",
   "num_attention_heads": 32,
   "num_hidden_layers": 48,
   "num_key_value_heads": 8,
   "rms_norm_eps": 1e-05,
-  "rope_theta": 10000.0,
-  "sliding_window": 4096,
   "tie_word_embeddings": false,
   "torch_dtype": "float16",
   "transformers_version": "4.35.2",

 {
+  "_name_or_path": "CallComply/Starling-LM-11B-alpha",
   "architectures": [
     "MistralForCausalLM"
   ],
   "hidden_size": 4096,
   "initializer_range": 0.02,
   "intermediate_size": 14336,
+  "max_position_embeddings": 32769,
   "model_type": "mistral",
   "num_attention_heads": 32,
   "num_hidden_layers": 48,
   "num_key_value_heads": 8,
   "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 100000.0,
+  "sliding_window": null,
   "tie_word_embeddings": false,
   "torch_dtype": "float16",
   "transformers_version": "4.35.2",