first commit

config.json

@@ -0,0 +1,46 @@
+{
+  "_name_or_path": "/data/bumengyu/3.work/lowResourceLLM/checkpoints/llama3-xbridge-nllb1.3B_enc_translation_y2en2x_10langs_3.6M-train_mapping_xattn-ot6.0_dec1.0-ot2_cosin-2e-5-then_general_50langs_2.5M-one_temp-train_encmapping-boundary-two_stage-2epo-4e-5",
+  "architectures": [
+    "LlamaForCasualLMWithXBridge"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dec_lambda": 1.0,
+  "eos_token_id": 2,
+  "freeze_dec": true,
+  "freeze_enc": true,
+  "freeze_llm": true,
+  "freeze_mapping_enc2llm": true,
+  "freeze_mapping_llm2dec": false,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "llm_bos_token_id": 128000,
+  "llm_eos_token_id": 128001,
+  "llm_pad_token_id": 128002,
+  "llm_path": "/data/bumengyu/2.model/llama3-8B",
+  "max_gen_len": 128,
+  "max_position_embeddings": 2048,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "mt_eos_token_id": 2,
+  "mt_pad_token_id": 1,
+  "mt_path": "/data/bumengyu/2.model/nllb-200-distilled-1.3B",
+  "mt_vocab_size": 256206,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "ot_lambda": 6.0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.2",
+  "use_cache": true,
+  "vocab_size": 32000
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.45.2"
+}

model-00001-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1207706703cf8a14129433d97f3b29c65678559a4b3c5b4ef9e6673562472f1
+size 4999871912

model-00002-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0a1a48f6cc89dac8ab4ee63c428e452460279747d6fe420b9c76cc62f25c297
+size 4966590720

model-00003-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0d9c2fc8fa086a432ce6a16eb2740fa5154f400657d4ce2f1ed3a1b8e624161
+size 4999813600

model-00004-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dab6bff6ac783579747c422c55ecfae0fd5e64609c46b8ed6546760065bd7a88
+size 4999813640

model-00005-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8ef36ebad04d3596c20dfdd5c61bd6628ba4b872d3ea4229c6e0fff68a186ac
+size 4832007968

model-00006-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ea2b01812e129bd2019e0137d891e2f37ac203cca1b6bb5c458cee895c4ab59
+size 4999813656

model-00007-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc773bee3e154584b2749f2aeeae8206fdcaddb8ff9268a39b7e7de3cc624900
+size 4999813656

model-00008-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:948f0f15c4336e99596182c5ce7a168932068f4590fa24737ab983eb7331347b
+size 3594779304