lunahr
/

SystemGemma2-2b-it

Text Generation

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

Piotr Zalewski commited on Aug 11

Commit

533c223

•

1 Parent(s): 86cac43

#2 retrying 32k context

Files changed (1) hide show

config.json +3 -3

config.json CHANGED Viewed

@@ -18,7 +18,7 @@
   "hidden_size": 2304,
   "initializer_range": 0.02,
   "intermediate_size": 9216,
-  "max_position_embeddings": 8192,
   "model_type": "gemma2",
   "num_attention_heads": 8,
   "num_hidden_layers": 26,
@@ -26,8 +26,8 @@
   "pad_token_id": 0,
   "query_pre_attn_scalar": 256,
   "rms_norm_eps": 1e-06,
-  "rope_theta": 10000.0,
-  "sliding_window": 4096,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.42.4",
   "use_cache": true,

   "hidden_size": 2304,
   "initializer_range": 0.02,
   "intermediate_size": 9216,
+  "max_position_embeddings": 32768,
   "model_type": "gemma2",
   "num_attention_heads": 8,
   "num_hidden_layers": 26,
   "pad_token_id": 0,
   "query_pre_attn_scalar": 256,
   "rms_norm_eps": 1e-06,
+  "rope_theta": 160000.0,
+  "sliding_window": 32768,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.42.4",
   "use_cache": true,