OpenPipe
/

Hermes-2-Theta-Llama-3-70B-32k

Text Generation

function calling

structured outputs

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

corbt commited on Jun 20

Commit

2a7d46d

•

1 Parent(s): 89f1ddb

extend context length to 32k

Files changed (1) hide show

config.json +2 -2

config.json CHANGED Viewed

@@ -11,7 +11,7 @@
   "hidden_size": 8192,
   "initializer_range": 0.02,
   "intermediate_size": 28672,
-  "max_position_embeddings": 8192,
   "mlp_bias": false,
   "model_type": "llama",
   "num_attention_heads": 64,
@@ -20,7 +20,7 @@
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,
-  "rope_theta": 500000.0,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.40.0.dev0",

   "hidden_size": 8192,
   "initializer_range": 0.02,
   "intermediate_size": 28672,
+  "max_position_embeddings": 32768,
   "mlp_bias": false,
   "model_type": "llama",
   "num_attention_heads": 64,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,
+  "rope_theta": 4000000.0,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.40.0.dev0",