End of training

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 0.9937888198757764,
+    "eval_loss": 1.4015706777572632,
+    "eval_runtime": 2.274,
+    "eval_samples": 518,
+    "eval_samples_per_second": 90.151,
+    "eval_steps_per_second": 0.88,
     "total_flos": 8.089924109389005e+17,
     "train_loss": 1.786448487639427,
     "train_runtime": 830.0773,

config.json

@@ -0,0 +1,32 @@
+{
+  "_attn_implementation_autoset": true,
+  "_name_or_path": "meta-llama/Llama-2-13b-hf",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 32000,
+  "eos_token_id": 32001,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 13824,
+  "max_position_embeddings": 4096,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 40,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 40,
+  "pad_token_id": 32001,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.2",
+  "use_cache": true,
+  "vocab_size": 32002
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.9937888198757764,
+    "eval_loss": 1.4015706777572632,
+    "eval_runtime": 2.274,
+    "eval_samples": 518,
+    "eval_samples_per_second": 90.151,
+    "eval_steps_per_second": 0.88
+}

runs/Nov22_17-00-23_dull-time-lies-fin-02/events.out.tfevents.1732295980.dull-time-lies-fin-02.48810.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fec22d2399b3fe04fc40e5554837bf0eba491454bbfac90e31af5cf5fac7585a
+size 354