Training in progress, step 500

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "HuggingFaceTB/SmolLM-135M",
+  "_name_or_path": "HuggingFaceTB/SmolLM-360M",
   "architectures": [
     "LlamaForCausalLM"
   ],
@@ -8,15 +8,15 @@
   "bos_token_id": 0,
   "eos_token_id": 0,
   "hidden_act": "silu",
-  "hidden_size": 576,
+  "hidden_size": 960,
   "initializer_range": 0.02,
-  "intermediate_size": 1536,
+  "intermediate_size": 2560,
   "max_position_embeddings": 2048,
   "mlp_bias": false,
   "model_type": "llama",
-  "num_attention_heads": 9,
-  "num_hidden_layers": 30,
-  "num_key_value_heads": 3,
+  "num_attention_heads": 15,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 5,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0966d884948bcaba341f5fa4a8388b39ba1e73003843f64d86e28eae3946294a
-size 269060552
+oid sha256:c9d7969370ec5ba5ff7d0a01009e561f5c7ee6121b11c443058964fee16ced85
+size 723674912

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a2e9aade2e84d2599ec223a3c6fc24b7e8b34e6efd714ef6c9d0af1689e5b796
+oid sha256:727195e2b84d4454eacb2515ab01cb9455537063ce8e6dc5d1f560d0999945bd
 size 5432