Upload BartForConditionalGeneration

config.json

@@ -0,0 +1,252 @@
+{
+  "_name_or_path": "ccdv/lsg-bart-large-4096",
+  "activation_dropout": 0.1,
+  "activation_function": "gelu",
+  "adaptive": true,
+  "add_bias_logits": false,
+  "add_final_layer_norm": false,
+  "architectures": [
+    "BartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.1,
+  "auto_map": {
+    "AutoConfig": "ccdv/lsg-bart-large-4096--modeling_lsg_bart.LSGBartConfig",
+    "AutoModel": "ccdv/lsg-bart-large-4096--modeling_lsg_bart.LSGBartModel",
+    "AutoModelForCausalLM": "ccdv/lsg-bart-large-4096--modeling_lsg_bart.LSGBartForCausalLM",
+    "AutoModelForQuestionAnswering": "ccdv/lsg-bart-large-4096--modeling_lsg_bart.LSGBartForQuestionAnswering",
+    "AutoModelForSeq2SeqLM": "ccdv/lsg-bart-large-4096--modeling_lsg_bart.LSGBartForConditionalGeneration",
+    "AutoModelForSequenceClassification": "ccdv/lsg-bart-large-4096--modeling_lsg_bart.LSGBartForSequenceClassification"
+  },
+  "base_model_prefix": "lsg",
+  "block_size": 128,
+  "bos_token_id": 0,
+  "classif_dropout": 0.1,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "early_stopping": true,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 0,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "lsh_num_pre_rounds": 1,
+  "mask_first_token": false,
+  "max_position_embeddings": 4096,
+  "model_type": "bart",
+  "no_repeat_ngram_size": 3,
+  "normalize_before": false,
+  "num_beams": 4,
+  "num_global_tokens": 1,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "pass_global_tokens_to_decoder": true,
+  "pool_with_global": true,
+  "quantization_config": {
+    "batch_size": 1,
+    "bits": 2,
+    "block_name_to_quantize": "model.decoder.layers",
+    "damp_percent": 0.1,
+    "dataset": "c4-new",
+    "desc_act": false,
+    "disable_exllama": false,
+    "group_size": 128,
+    "model_seqlen": 4096,
+    "module_name_preceding_first_block": [
+      "model.shared",
+      "model.encoder.embed_tokens",
+      "model.encoder.embed_positions",
+      "model.encoder.layers.0.self_attn.k_proj",
+      "model.encoder.layers.0.self_attn.v_proj",
+      "model.encoder.layers.0.self_attn.q_proj",
+      "model.encoder.layers.0.self_attn.out_proj",
+      "model.encoder.layers.0.self_attn",
+      "model.encoder.layers.0.self_attn_layer_norm",
+      "model.encoder.layers.0.activation_fn",
+      "model.encoder.layers.0.fc1",
+      "model.encoder.layers.0.fc2",
+      "model.encoder.layers.0.final_layer_norm",
+      "model.encoder.layers.0",
+      "model.encoder.layers.1.self_attn.k_proj",
+      "model.encoder.layers.1.self_attn.v_proj",
+      "model.encoder.layers.1.self_attn.q_proj",
+      "model.encoder.layers.1.self_attn.out_proj",
+      "model.encoder.layers.1.self_attn",
+      "model.encoder.layers.1.self_attn_layer_norm",
+      "model.encoder.layers.1.activation_fn",
+      "model.encoder.layers.1.fc1",
+      "model.encoder.layers.1.fc2",
+      "model.encoder.layers.1.final_layer_norm",
+      "model.encoder.layers.1",
+      "model.encoder.layers.2.self_attn.k_proj",
+      "model.encoder.layers.2.self_attn.v_proj",
+      "model.encoder.layers.2.self_attn.q_proj",
+      "model.encoder.layers.2.self_attn.out_proj",
+      "model.encoder.layers.2.self_attn",
+      "model.encoder.layers.2.self_attn_layer_norm",
+      "model.encoder.layers.2.activation_fn",
+      "model.encoder.layers.2.fc1",
+      "model.encoder.layers.2.fc2",
+      "model.encoder.layers.2.final_layer_norm",
+      "model.encoder.layers.2",
+      "model.encoder.layers.3.self_attn.k_proj",
+      "model.encoder.layers.3.self_attn.v_proj",
+      "model.encoder.layers.3.self_attn.q_proj",
+      "model.encoder.layers.3.self_attn.out_proj",
+      "model.encoder.layers.3.self_attn",
+      "model.encoder.layers.3.self_attn_layer_norm",
+      "model.encoder.layers.3.activation_fn",
+      "model.encoder.layers.3.fc1",
+      "model.encoder.layers.3.fc2",
+      "model.encoder.layers.3.final_layer_norm",
+      "model.encoder.layers.3",
+      "model.encoder.layers.4.self_attn.k_proj",
+      "model.encoder.layers.4.self_attn.v_proj",
+      "model.encoder.layers.4.self_attn.q_proj",
+      "model.encoder.layers.4.self_attn.out_proj",
+      "model.encoder.layers.4.self_attn",
+      "model.encoder.layers.4.self_attn_layer_norm",
+      "model.encoder.layers.4.activation_fn",
+      "model.encoder.layers.4.fc1",
+      "model.encoder.layers.4.fc2",
+      "model.encoder.layers.4.final_layer_norm",
+      "model.encoder.layers.4",
+      "model.encoder.layers.5.self_attn.k_proj",
+      "model.encoder.layers.5.self_attn.v_proj",
+      "model.encoder.layers.5.self_attn.q_proj",
+      "model.encoder.layers.5.self_attn.out_proj",
+      "model.encoder.layers.5.self_attn",
+      "model.encoder.layers.5.self_attn_layer_norm",
+      "model.encoder.layers.5.activation_fn",
+      "model.encoder.layers.5.fc1",
+      "model.encoder.layers.5.fc2",
+      "model.encoder.layers.5.final_layer_norm",
+      "model.encoder.layers.5",
+      "model.encoder.layers.6.self_attn.k_proj",
+      "model.encoder.layers.6.self_attn.v_proj",
+      "model.encoder.layers.6.self_attn.q_proj",
+      "model.encoder.layers.6.self_attn.out_proj",
+      "model.encoder.layers.6.self_attn",
+      "model.encoder.layers.6.self_attn_layer_norm",
+      "model.encoder.layers.6.activation_fn",
+      "model.encoder.layers.6.fc1",
+      "model.encoder.layers.6.fc2",
+      "model.encoder.layers.6.final_layer_norm",
+      "model.encoder.layers.6",
+      "model.encoder.layers.7.self_attn.k_proj",
+      "model.encoder.layers.7.self_attn.v_proj",
+      "model.encoder.layers.7.self_attn.q_proj",
+      "model.encoder.layers.7.self_attn.out_proj",
+      "model.encoder.layers.7.self_attn",
+      "model.encoder.layers.7.self_attn_layer_norm",
+      "model.encoder.layers.7.activation_fn",
+      "model.encoder.layers.7.fc1",
+      "model.encoder.layers.7.fc2",
+      "model.encoder.layers.7.final_layer_norm",
+      "model.encoder.layers.7",
+      "model.encoder.layers.8.self_attn.k_proj",
+      "model.encoder.layers.8.self_attn.v_proj",
+      "model.encoder.layers.8.self_attn.q_proj",
+      "model.encoder.layers.8.self_attn.out_proj",
+      "model.encoder.layers.8.self_attn",
+      "model.encoder.layers.8.self_attn_layer_norm",
+      "model.encoder.layers.8.activation_fn",
+      "model.encoder.layers.8.fc1",
+      "model.encoder.layers.8.fc2",
+      "model.encoder.layers.8.final_layer_norm",
+      "model.encoder.layers.8",
+      "model.encoder.layers.9.self_attn.k_proj",
+      "model.encoder.layers.9.self_attn.v_proj",
+      "model.encoder.layers.9.self_attn.q_proj",
+      "model.encoder.layers.9.self_attn.out_proj",
+      "model.encoder.layers.9.self_attn",
+      "model.encoder.layers.9.self_attn_layer_norm",
+      "model.encoder.layers.9.activation_fn",
+      "model.encoder.layers.9.fc1",
+      "model.encoder.layers.9.fc2",
+      "model.encoder.layers.9.final_layer_norm",
+      "model.encoder.layers.9",
+      "model.encoder.layers.10.self_attn.k_proj",
+      "model.encoder.layers.10.self_attn.v_proj",
+      "model.encoder.layers.10.self_attn.q_proj",
+      "model.encoder.layers.10.self_attn.out_proj",
+      "model.encoder.layers.10.self_attn",
+      "model.encoder.layers.10.self_attn_layer_norm",
+      "model.encoder.layers.10.activation_fn",
+      "model.encoder.layers.10.fc1",
+      "model.encoder.layers.10.fc2",
+      "model.encoder.layers.10.final_layer_norm",
+      "model.encoder.layers.10",
+      "model.encoder.layers.11.self_attn.k_proj",
+      "model.encoder.layers.11.self_attn.v_proj",
+      "model.encoder.layers.11.self_attn.q_proj",
+      "model.encoder.layers.11.self_attn.out_proj",
+      "model.encoder.layers.11.self_attn",
+      "model.encoder.layers.11.self_attn_layer_norm",
+      "model.encoder.layers.11.activation_fn",
+      "model.encoder.layers.11.fc1",
+      "model.encoder.layers.11.fc2",
+      "model.encoder.layers.11.final_layer_norm",
+      "model.encoder.layers.11",
+      "model.encoder.layers",
+      "model.encoder.layernorm_embedding",
+      "model.encoder",
+      "model.decoder.embed_tokens",
+      "model.decoder.embed_positions"
+    ],
+    "pad_token_id": null,
+    "quant_method": "gptq",
+    "sym": true,
+    "tokenizer": null,
+    "true_sequential": true,
+    "use_cuda_fp16": true
+  },
+  "scale_embedding": false,
+  "sparse_block_size": 128,
+  "sparsity_factor": 2,
+  "sparsity_type": "norm",
+  "task_specific_params": {
+    "summarization": {
+      "length_penalty": 1.0,
+      "max_length": 128,
+      "min_length": 12,
+      "num_beams": 4
+    },
+    "summarization_cnn": {
+      "length_penalty": 2.0,
+      "max_length": 142,
+      "min_length": 56,
+      "num_beams": 4
+    },
+    "summarization_xsum": {
+      "length_penalty": 1.0,
+      "max_length": 62,
+      "min_length": 11,
+      "num_beams": 6
+    }
+  },
+  "torch_dtype": "float16",
+  "transformers_version": "4.33.2",
+  "use_cache": true,
+  "vocab_size": 50265
+}

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "early_stopping": true,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 0,
+  "forced_eos_token_id": 2,
+  "no_repeat_ngram_size": 3,
+  "num_beams": 4,
+  "pad_token_id": 1,
+  "transformers_version": "4.33.2"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf3902039942cf09cb50810fd6d3ce995d8d8f05e4d7cb51074a854ae6d96ab5
+size 683307189