Upload folder using huggingface_hub

config.json

@@ -1,87 +1,34 @@
 {
-  "vocab_size": 28776,
-  "d_model": 1024,
-  "d_kv": 64,
+  "_name_or_path": "pszemraj/tFINE-850m-24x24-1024ctx",
+  "architectures": [
+    "T5ForConditionalGeneration"
+  ],
+  "classifier_dropout": 0.0,
   "d_ff": 3072,
-  "num_layers": 16,
+  "d_kv": 64,
+  "d_model": 1024,
+  "decoder_start_token_id": 3,
+  "dense_act_fn": "silu",
+  "dropout_rate": 0.0,
+  "eos_token_id": 2,
+  "feed_forward_proj": "gated-silu",
+  "initializer_factor": 1.0,
+  "is_bf16": true,
+  "is_encoder_decoder": false,
+  "is_gated_act": true,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "t5",
   "num_decoder_layers": 16,
   "num_heads": 16,
   "num_key_value_heads": 4,
-  "relative_attention_num_buckets": 48,
+  "num_layers": 16,
+  "output_past": true,
+  "pad_token_id": 3,
   "relative_attention_max_distance": 128,
-  "dropout_rate": 0.0,
-  "classifier_dropout": 0.0,
-  "layer_norm_epsilon": 1e-06,
-  "initializer_factor": 1.0,
-  "feed_forward_proj": "gated-silu",
-  "use_cache": true,
-  "use_gqa": true,
-  "dense_act_fn": "silu",
-  "is_gated_act": true,
-  "return_dict": true,
-  "output_hidden_states": false,
-  "output_attentions": false,
-  "torchscript": false,
-  "torch_dtype": null,
-  "use_bfloat16": false,
-  "tf_legacy_loss": false,
-  "pruned_heads": {},
+  "relative_attention_num_buckets": 48,
   "tie_word_embeddings": false,
-  "chunk_size_feed_forward": 0,
-  "is_encoder_decoder": true,
-  "is_decoder": false,
-  "cross_attention_hidden_size": null,
-  "add_cross_attention": false,
-  "tie_encoder_decoder": false,
-  "max_length": 20,
-  "min_length": 0,
-  "do_sample": false,
-  "early_stopping": false,
-  "num_beams": 1,
-  "num_beam_groups": 1,
-  "diversity_penalty": 0.0,
-  "temperature": 1.0,
-  "top_k": 50,
-  "top_p": 1.0,
-  "typical_p": 1.0,
-  "repetition_penalty": 1.0,
-  "length_penalty": 1.0,
-  "no_repeat_ngram_size": 0,
-  "encoder_no_repeat_ngram_size": 0,
-  "bad_words_ids": null,
-  "num_return_sequences": 1,
-  "output_scores": false,
-  "return_dict_in_generate": false,
-  "forced_bos_token_id": null,
-  "forced_eos_token_id": null,
-  "remove_invalid_values": false,
-  "exponential_decay_length_penalty": null,
-  "suppress_tokens": null,
-  "begin_suppress_tokens": null,
-  "architectures": [
-    "T5ForConditionalGeneration"
-  ],
-  "finetuning_task": null,
-  "id2label": {
-    "0": "LABEL_0",
-    "1": "LABEL_1"
-  },
-  "label2id": {
-    "LABEL_0": 0,
-    "LABEL_1": 1
-  },
-  "tokenizer_class": null,
-  "prefix": null,
-  "bos_token_id": null,
-  "pad_token_id": 3,
-  "eos_token_id": 2,
-  "sep_token_id": null,
-  "decoder_start_token_id": 3,
-  "task_specific_params": null,
-  "problem_type": null,
-  "_name_or_path": "/workspace/nanoT5/logs/2024-10-20/18-25-17/checkpoint-pt-27000",
   "transformers_version": "4.46.0.dev0",
-  "is_bf16": true,
-  "model_type": "t5",
-  "output_past": true
-}
+  "use_cache": true,
+  "use_gqa": true,
+  "vocab_size": 28776
+}

test.py

@@ -6,8 +6,8 @@ tokenizer = AutoTokenizer.from_pretrained("BEE-spoke-data/hf_slimpajama-6B-28672
 special_tokens_dict = {'additional_special_tokens': ['[R]', '[S]', '[X]', '[NTP]']}
 tokenizer.add_special_tokens(special_tokens_dict)
 
-model = AutoModelForSeq2SeqLM.from_pretrained("/workspace/nanoT5/logs/2024-10-20/18-25-17/checkpoint-pt-27000").to("cuda")
-prompt = "The "
+model = AutoModelForSeq2SeqLM.from_pretrained("/workspace/nanoT5/logs/2024-10-20/18-25-17/amazingvince/ul3-base").to("cuda")
+prompt = "[NTP] The "
 inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
 # Add decoder_input_ids
 # decoder_input_ids = torch.ones((inputs.input_ids.shape[0], 1), dtype=torch.long) * model.config.decoder_start_token_id