tangledgroup
/

tangled-alpha-0.11-core

Text Generation

Model card Files Files and versions Community

mtasic85 commited on 20 days ago

Commit

aa44061

·

1 Parent(s): 26a07d4

pretrain

Files changed (1) hide show

config-0.json +4 -4

config-0.json CHANGED Viewed

@@ -8,15 +8,15 @@
   "eos_token_id": 1,
   "head_dim": 64,
   "hidden_act": "silu",
-  "hidden_size": 768,
   "initializer_range": 0.02,
-  "intermediate_size": 2048,
   "max_position_embeddings": 131072,
   "mlp_bias": false,
   "model_type": "llama",
-  "num_attention_heads": 12,
   "num_hidden_layers": 32,
-  "num_key_value_heads": 4,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,

   "eos_token_id": 1,
   "head_dim": 64,
   "hidden_act": "silu",
+  "hidden_size": 512,
   "initializer_range": 0.02,
+  "intermediate_size": 1365,
   "max_position_embeddings": 131072,
   "mlp_bias": false,
   "model_type": "llama",
+  "num_attention_heads": 8,
   "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,