LLM360
/

K2

Text Generation

Inference Endpoints

text-generation-inference

Model card Files Files and versions Community

yukiontheiceberg commited on 3 days ago

Commit

ef6007e

•

1 Parent(s): c3e2f4f

Update config.json

Files changed (1) hide show

config.json +3 -3

config.json CHANGED Viewed

@@ -11,8 +11,8 @@
   "hidden_size": 8192,
   "initializer_range": 0.02,
   "intermediate_size": 22016,
-  "max_position_embeddings": 2048,
-  "max_sequence_length": 2048,
   "model_type": "llama",
   "num_attention_heads": 64,
   "num_hidden_layers": 80,
@@ -21,7 +21,7 @@
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,
-  "rope_theta": 10000.0,
   "tie_word_embeddings": false,
   "torch_dtype": "float16",
   "transformers_version": "4.39.3",

   "hidden_size": 8192,
   "initializer_range": 0.02,
   "intermediate_size": 22016,
+  "max_position_embeddings": 8192,
+  "max_sequence_length": 8192,
   "model_type": "llama",
   "num_attention_heads": 64,
   "num_hidden_layers": 80,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,
+  "rope_theta": 500000.0,
   "tie_word_embeddings": false,
   "torch_dtype": "float16",
   "transformers_version": "4.39.3",