Training in progress, step 4000

config.json

@@ -1,28 +1,32 @@
 {
+  "activation_function": "gelu_new",
   "architectures": [
-    "RobertaForMaskedLM"
+    "GPT2LMHeadModel"
   ],
-  "attention_probs_dropout_prob": 0.1,
-  "bos_token_id": 3,
-  "classifier_dropout": null,
-  "eos_token_id": 4,
-  "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.1,
-  "hidden_size": 256,
+  "attn_pdrop": 0.1,
+  "bos_token_id": 0,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 1,
   "initializer_range": 0.02,
-  "intermediate_size": 2048,
-  "layer_norm_eps": 1e-05,
-  "mask_token_id": 6,
-  "max_position_embeddings": 512,
-  "model_type": "roberta",
-  "num_attention_heads": 8,
-  "num_hidden_layers": 8,
-  "pad_token_id": 1,
-  "position_embedding_type": "absolute",
-  "tie_word_embeddings": false,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 512,
+  "n_embd": 256,
+  "n_head": 8,
+  "n_inner": 2048,
+  "n_layer": 8,
+  "n_positions": 512,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
   "torch_dtype": "float32",
   "transformers_version": "4.45.2",
-  "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 8192
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2b753ba321eda5063cfa7df56a7ef1cdc691c4014810d92ec26ac26b03ad887
-size 59702184
+oid sha256:2801ae148db1f89a38af1d83271bf1f654930608ffbdc05c9e66f5d71a959394
+size 51007160

special_tokens_map.json

@@ -1,6 +1,34 @@
 {
-  "bos_token": "[PAD]",
-  "eos_token": "[EOS]",
-  "mask_token": "[MASK]",
-  "pad_token": "[CLS]"
+  "additional_special_tokens": [
+    "<|endoftext|>"
+  ],
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
 }

tokenizer_config.json

@@ -1,7 +1,7 @@
 {
   "added_tokens_decoder": {
     "0": {
-      "content": "[UNK]",
+      "content": "<s>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -9,7 +9,7 @@
       "special": true
     },
     "1": {
-      "content": "[CLS]",
+      "content": "</s>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -17,7 +17,7 @@
       "special": true
     },
     "2": {
-      "content": "[SEP]",
+      "content": "<pad>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -25,7 +25,7 @@
       "special": true
     },
     "3": {
-      "content": "[PAD]",
+      "content": "<unk>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -33,7 +33,7 @@
       "special": true
     },
     "4": {
-      "content": "[EOS]",
+      "content": "<mask>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -41,15 +41,7 @@
       "special": true
     },
     "5": {
-      "content": "[BOS]",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "6": {
-      "content": "[MASK]",
+      "content": "<|endoftext|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -57,11 +49,15 @@
       "special": true
     }
   },
-  "bos_token": "[PAD]",
+  "additional_special_tokens": [
+    "<|endoftext|>"
+  ],
+  "bos_token": "<s>",
   "clean_up_tokenization_spaces": false,
-  "eos_token": "[EOS]",
-  "mask_token": "[MASK]",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
   "model_max_length": 128,
-  "pad_token": "[CLS]",
-  "tokenizer_class": "PreTrainedTokenizerFast"
+  "pad_token": "</s>",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82b00b33ff5a34cb27834747d142985f370938cb613e5baed1026ce1070e9a42
+oid sha256:2a77d647bb2e817d836b30fa0c2702de6364acb70c98f3c308390594352b8543
 size 5368