Checkpoint at step 60000

checkpoint_60000/config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "EleutherAI/pythia-160M",
+  "_name_or_path": "EleutherAI/pythia-14M",
   "architectures": [
     "GPTNeoXForCausalLM"
   ],
@@ -10,14 +10,14 @@
   "eos_token_id": 0,
   "hidden_act": "gelu",
   "hidden_dropout": 0.0,
-  "hidden_size": 768,
+  "hidden_size": 128,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 512,
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 2048,
   "model_type": "gpt_neox",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "num_attention_heads": 4,
+  "num_hidden_layers": 6,
   "output_hidden_states": true,
   "partial_rotary_factor": 0.25,
   "rope_scaling": null,

checkpoint_60000/generation_config.json

@@ -2,7 +2,5 @@
   "_from_model_config": true,
   "bos_token_id": 0,
   "eos_token_id": 0,
-  "output_hidden_states": true,
-  "return_dict_in_generate": true,
   "transformers_version": "4.45.1"
 }

checkpoint_60000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be75cbd7e466a72a7b633e90071ce563b8133015b70140e035b75aa48027f70b
-size 649173560
+oid sha256:1973bc922c72998cacf231be9d952976b07c1bd27db7dd18c60776b0533b8a74
+size 56256816