Training in progress, step 6000

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "lectura/TinyLlama-120M-news",
+  "_name_or_path": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
   "architectures": [
     "BitLlamaForCausalLM"
   ],
@@ -9,21 +9,21 @@
   "bos_token_id": 1,
   "eos_token_id": 2,
   "hidden_act": "silu",
-  "hidden_size": 768,
+  "hidden_size": 2048,
   "initializer_range": 0.02,
-  "intermediate_size": 2048,
-  "max_position_embeddings": 1024,
+  "intermediate_size": 5632,
+  "max_position_embeddings": 2048,
   "model_type": "bit_llama",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
-  "num_key_value_heads": 12,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 22,
+  "num_key_value_heads": 4,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,
   "rope_theta": 10000.0,
-  "tie_word_embeddings": true,
+  "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.39.0.dev0",
   "use_cache": false,
-  "vocab_size": 46338
+  "vocab_size": 32002
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82082c0b5de2f82febe83a84ec395efa8ad3d6d19b52c41268d202c52dd39484
-size 241437456
+oid sha256:9b52612851e8d6aa212ce306d9e797a5a7e6b359edc5e9db3ff60568afca1b2d
+size 2201749928

tokenizer_config.json

@@ -26,47 +26,7 @@
       "single_word": false,
       "special": true
     },
-    "46331": {
-      "content": "<|sep|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "46332": {
-      "content": "<|endoftext|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "46333": {
-      "content": "<|acc|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "46334": {
-      "content": "<|rrn|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "46335": {
-      "content": "<|tel|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "46336": {
+    "32000": {
       "content": "<|im_start|>",
       "lstrip": false,
       "normalized": false,
@@ -74,7 +34,7 @@
       "single_word": false,
       "special": true
     },
-    "46337": {
+    "32001": {
       "content": "<|im_end|>",
       "lstrip": false,
       "normalized": false,
@@ -94,8 +54,9 @@
   "legacy": false,
   "model_max_length": 2048,
   "pad_token": "<|im_end|>",
+  "padding_side": "right",
   "sp_model_kwargs": {},
   "tokenizer_class": "LlamaTokenizer",
   "unk_token": "<unk>",
-  "use_default_system_prompt": true
+  "use_default_system_prompt": false
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:835feb71d544b963e5d2bc0c5bc229d8fdfc8cb508e797df572da8dba67d19f9
+oid sha256:07cc40c5a6d24066e3c9c9dc23db4ba29b9036550b49d4e3f00420e402d9f406
 size 4984