Upload folder using huggingface_hub

README.md

@@ -0,0 +1,55 @@
+---
+tags:
+- merge
+- mergekit
+- lazymergekit
+- tiiuae/falcon-11B
+base_model:
+- tiiuae/falcon-11B
+- tiiuae/falcon-11B
+---
+
+# Falcon2-5.5B-Romanian
+
+Falcon2-5.5B-Romanian is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
+* [tiiuae/falcon-11B](https://huggingface.co/tiiuae/falcon-11B)
+* [tiiuae/falcon-11B](https://huggingface.co/tiiuae/falcon-11B)
+
+## 🧩 Configuration
+
+```yaml
+slices:
+  - sources:
+      - model: tiiuae/falcon-11B
+        layer_range: [0, 25]
+  - sources:
+      - model: tiiuae/falcon-11B
+        layer_range: [56, 59]
+merge_method: passthrough
+dtype: bfloat16
+```
+
+## 💻 Usage
+
+```python
+!pip install -qU transformers accelerate
+
+from transformers import AutoTokenizer
+import transformers
+import torch
+
+model = "ssmits/Falcon2-5.5B-Romanian"
+messages = [{"role": "user", "content": "What is a large language model?"}]
+
+tokenizer = AutoTokenizer.from_pretrained(model)
+prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+pipeline = transformers.pipeline(
+    "text-generation",
+    model=model,
+    torch_dtype=torch.float16,
+    device_map="auto",
+)
+
+outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
+print(outputs[0]["generated_text"])
+```

config.json

@@ -0,0 +1,42 @@
+{
+  "_name_or_path": "tiiuae/falcon-11B",
+  "activation": "gelu",
+  "alibi": false,
+  "architectures": [
+    "FalconForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "tiiuae/falcon-11B--configuration_falcon.FalconConfig",
+    "AutoModel": "tiiuae/falcon-11B--modeling_falcon.FalconModel",
+    "AutoModelForCausalLM": "tiiuae/falcon-11B--modeling_falcon.FalconForCausalLM",
+    "AutoModelForQuestionAnswering": "tiiuae/falcon-11B--modeling_falcon.FalconForQuestionAnswering",
+    "AutoModelForSequenceClassification": "tiiuae/falcon-11B--modeling_falcon.FalconForSequenceClassification",
+    "AutoModelForTokenClassification": "tiiuae/falcon-11B--modeling_falcon.FalconForTokenClassification"
+  },
+  "bias": false,
+  "bos_token_id": 11,
+  "eos_token_id": 11,
+  "ff_factor": 4,
+  "ffn_hidden_size": 16384,
+  "hidden_dropout": 0.0,
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "max_position_embeddings": 8192,
+  "model_type": "falcon",
+  "multi_query": true,
+  "new_decoder_architecture": true,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 28,
+  "num_kv_heads": 8,
+  "num_ln_in_parallel_attn": 1,
+  "parallel_attn": true,
+  "rope_scaling": null,
+  "rope_theta": 500042.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.2",
+  "use_cache": true,
+  "vocab_size": 65024
+}

mergekit_config.yml

@@ -0,0 +1,10 @@
+
+slices:
+  - sources:
+      - model: tiiuae/falcon-11B
+        layer_range: [0, 25]
+  - sources:
+      - model: tiiuae/falcon-11B
+        layer_range: [56, 59]
+merge_method: passthrough
+dtype: bfloat16

model-00001-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be473743446f6310af51e0c376f85362497f956b299ae82088be850871ebe058
+size 885031960

model-00002-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7958aad384a953d794c245dc1fb049cf2e88db6555ebef8a37db39727a4f268
+size 973112984

model-00003-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:742281f7d38f9ea479d6ad1458904c57428323cfe147476a9fc881340d3ea5a2
+size 922798072

model-00004-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bf29decce4e70a2158383597b1821eece07e06f1f58d629d62a30f47ce30bb2
+size 922798080

model-00005-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1926ea5d345d3aefefe565f668d2c1e61e09bc21ed0a3e2b7ef371643a55deed
+size 973112992

model-00006-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:612dc205bbe5d2e8b06a9dcd4c34aeb7bd041b59786defea49fe20c62ad08759
+size 922798064

model-00007-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b49e0a01cb4bf426c6781a0018642b2b0fb903ddfa45c2f1cc922015b4f0ebc
+size 922798080

model-00008-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46663363935d36efe732f818a0189aa38f01fd34c466b8162870cb8913c0d5e1
+size 973112976

model-00009-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a997ff817469d17ed339b301fd69005a6cc071bddb9210e041cf868572fd22fd
+size 922798056

model-00010-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fdc8f6860b3cc7296be0bdbc6e39218c54003d371ccb211394aa64d0fb3d7b3
+size 922798072

model-00011-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85e48d70ad97bd2153b3a7c46b69fe2e65fa58067a4946e66a267810d0044857
+size 973112976

model-00012-of-00012.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c287b5e845465167828d1374df8f4ccfa69a9d26eed0d02a26c7afb2b22e5e4a
+size 616579648

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.4.2", "total_size": 10930831360}, "weight_map": {"lm_head.weight": "model-00001-of-00012.safetensors", "transformer.h.0.input_layernorm.bias": "model-00001-of-00012.safetensors", "transformer.h.0.input_layernorm.weight": "model-00001-of-00012.safetensors", "transformer.h.0.mlp.dense_4h_to_h.weight": "model-00001-of-00012.safetensors", "transformer.h.0.mlp.dense_h_to_4h.weight": "model-00001-of-00012.safetensors", "transformer.h.0.self_attention.dense.weight": "model-00001-of-00012.safetensors", "transformer.h.0.self_attention.query_key_value.weight": "model-00001-of-00012.safetensors", "transformer.h.1.input_layernorm.bias": "model-00001-of-00012.safetensors", "transformer.h.1.input_layernorm.weight": "model-00001-of-00012.safetensors", "transformer.h.1.mlp.dense_4h_to_h.weight": "model-00002-of-00012.safetensors", "transformer.h.1.mlp.dense_h_to_4h.weight": "model-00002-of-00012.safetensors", "transformer.h.1.self_attention.dense.weight": "model-00002-of-00012.safetensors", "transformer.h.1.self_attention.query_key_value.weight": "model-00002-of-00012.safetensors", "transformer.h.10.input_layernorm.bias": "model-00002-of-00012.safetensors", "transformer.h.10.input_layernorm.weight": "model-00002-of-00012.safetensors", "transformer.h.10.mlp.dense_4h_to_h.weight": "model-00002-of-00012.safetensors", "transformer.h.10.mlp.dense_h_to_4h.weight": "model-00002-of-00012.safetensors", "transformer.h.10.self_attention.dense.weight": "model-00002-of-00012.safetensors", "transformer.h.10.self_attention.query_key_value.weight": "model-00002-of-00012.safetensors", "transformer.h.11.input_layernorm.bias": "model-00002-of-00012.safetensors", "transformer.h.11.input_layernorm.weight": "model-00002-of-00012.safetensors", "transformer.h.11.mlp.dense_4h_to_h.weight": "model-00002-of-00012.safetensors", "transformer.h.11.mlp.dense_h_to_4h.weight": "model-00002-of-00012.safetensors", "transformer.h.11.self_attention.dense.weight": "model-00003-of-00012.safetensors", "transformer.h.11.self_attention.query_key_value.weight": "model-00003-of-00012.safetensors", "transformer.h.12.input_layernorm.bias": "model-00003-of-00012.safetensors", "transformer.h.12.input_layernorm.weight": "model-00003-of-00012.safetensors", "transformer.h.12.mlp.dense_4h_to_h.weight": "model-00003-of-00012.safetensors", "transformer.h.12.mlp.dense_h_to_4h.weight": "model-00003-of-00012.safetensors", "transformer.h.12.self_attention.dense.weight": "model-00003-of-00012.safetensors", "transformer.h.12.self_attention.query_key_value.weight": "model-00003-of-00012.safetensors", "transformer.h.13.input_layernorm.bias": "model-00003-of-00012.safetensors", "transformer.h.13.input_layernorm.weight": "model-00003-of-00012.safetensors", "transformer.h.13.mlp.dense_4h_to_h.weight": "model-00003-of-00012.safetensors", "transformer.h.13.mlp.dense_h_to_4h.weight": "model-00003-of-00012.safetensors", "transformer.h.13.self_attention.dense.weight": "model-00003-of-00012.safetensors", "transformer.h.13.self_attention.query_key_value.weight": "model-00003-of-00012.safetensors", "transformer.h.14.input_layernorm.bias": "model-00003-of-00012.safetensors", "transformer.h.14.input_layernorm.weight": "model-00003-of-00012.safetensors", "transformer.h.14.mlp.dense_4h_to_h.weight": "model-00003-of-00012.safetensors", "transformer.h.14.mlp.dense_h_to_4h.weight": "model-00004-of-00012.safetensors", "transformer.h.14.self_attention.dense.weight": "model-00004-of-00012.safetensors", "transformer.h.14.self_attention.query_key_value.weight": "model-00004-of-00012.safetensors", "transformer.h.15.input_layernorm.bias": "model-00004-of-00012.safetensors", "transformer.h.15.input_layernorm.weight": "model-00004-of-00012.safetensors", "transformer.h.15.mlp.dense_4h_to_h.weight": "model-00004-of-00012.safetensors", "transformer.h.15.mlp.dense_h_to_4h.weight": "model-00004-of-00012.safetensors", "transformer.h.15.self_attention.dense.weight": "model-00004-of-00012.safetensors", "transformer.h.15.self_attention.query_key_value.weight": "model-00004-of-00012.safetensors", "transformer.h.16.input_layernorm.bias": "model-00004-of-00012.safetensors", "transformer.h.16.input_layernorm.weight": "model-00004-of-00012.safetensors", "transformer.h.16.mlp.dense_4h_to_h.weight": "model-00004-of-00012.safetensors", "transformer.h.16.mlp.dense_h_to_4h.weight": "model-00004-of-00012.safetensors", "transformer.h.16.self_attention.dense.weight": "model-00004-of-00012.safetensors", "transformer.h.16.self_attention.query_key_value.weight": "model-00004-of-00012.safetensors", "transformer.h.17.input_layernorm.bias": "model-00004-of-00012.safetensors", "transformer.h.17.input_layernorm.weight": "model-00004-of-00012.safetensors", "transformer.h.17.mlp.dense_4h_to_h.weight": "model-00005-of-00012.safetensors", "transformer.h.17.mlp.dense_h_to_4h.weight": "model-00005-of-00012.safetensors", "transformer.h.17.self_attention.dense.weight": "model-00005-of-00012.safetensors", "transformer.h.17.self_attention.query_key_value.weight": "model-00005-of-00012.safetensors", "transformer.h.18.input_layernorm.bias": "model-00005-of-00012.safetensors", "transformer.h.18.input_layernorm.weight": "model-00005-of-00012.safetensors", "transformer.h.18.mlp.dense_4h_to_h.weight": "model-00005-of-00012.safetensors", "transformer.h.18.mlp.dense_h_to_4h.weight": "model-00005-of-00012.safetensors", "transformer.h.18.self_attention.dense.weight": "model-00005-of-00012.safetensors", "transformer.h.18.self_attention.query_key_value.weight": "model-00005-of-00012.safetensors", "transformer.h.19.input_layernorm.bias": "model-00005-of-00012.safetensors", "transformer.h.19.input_layernorm.weight": "model-00005-of-00012.safetensors", "transformer.h.19.mlp.dense_4h_to_h.weight": "model-00005-of-00012.safetensors", "transformer.h.19.mlp.dense_h_to_4h.weight": "model-00005-of-00012.safetensors", "transformer.h.19.self_attention.dense.weight": "model-00006-of-00012.safetensors", "transformer.h.19.self_attention.query_key_value.weight": "model-00006-of-00012.safetensors", "transformer.h.2.input_layernorm.bias": "model-00006-of-00012.safetensors", "transformer.h.2.input_layernorm.weight": "model-00006-of-00012.safetensors", "transformer.h.2.mlp.dense_4h_to_h.weight": "model-00006-of-00012.safetensors", "transformer.h.2.mlp.dense_h_to_4h.weight": "model-00006-of-00012.safetensors", "transformer.h.2.self_attention.dense.weight": "model-00006-of-00012.safetensors", "transformer.h.2.self_attention.query_key_value.weight": "model-00006-of-00012.safetensors", "transformer.h.20.input_layernorm.bias": "model-00006-of-00012.safetensors", "transformer.h.20.input_layernorm.weight": "model-00006-of-00012.safetensors", "transformer.h.20.mlp.dense_4h_to_h.weight": "model-00006-of-00012.safetensors", "transformer.h.20.mlp.dense_h_to_4h.weight": "model-00006-of-00012.safetensors", "transformer.h.20.self_attention.dense.weight": "model-00006-of-00012.safetensors", "transformer.h.20.self_attention.query_key_value.weight": "model-00006-of-00012.safetensors", "transformer.h.21.input_layernorm.bias": "model-00006-of-00012.safetensors", "transformer.h.21.input_layernorm.weight": "model-00006-of-00012.safetensors", "transformer.h.21.mlp.dense_4h_to_h.weight": "model-00006-of-00012.safetensors", "transformer.h.21.mlp.dense_h_to_4h.weight": "model-00007-of-00012.safetensors", "transformer.h.21.self_attention.dense.weight": "model-00007-of-00012.safetensors", "transformer.h.21.self_attention.query_key_value.weight": "model-00007-of-00012.safetensors", "transformer.h.22.input_layernorm.bias": "model-00007-of-00012.safetensors", "transformer.h.22.input_layernorm.weight": "model-00007-of-00012.safetensors", "transformer.h.22.mlp.dense_4h_to_h.weight": "model-00007-of-00012.safetensors", "transformer.h.22.mlp.dense_h_to_4h.weight": "model-00007-of-00012.safetensors", "transformer.h.22.self_attention.dense.weight": "model-00007-of-00012.safetensors", "transformer.h.22.self_attention.query_key_value.weight": "model-00007-of-00012.safetensors", "transformer.h.23.input_layernorm.bias": "model-00007-of-00012.safetensors", "transformer.h.23.input_layernorm.weight": "model-00007-of-00012.safetensors", "transformer.h.23.mlp.dense_4h_to_h.weight": "model-00007-of-00012.safetensors", "transformer.h.23.mlp.dense_h_to_4h.weight": "model-00007-of-00012.safetensors", "transformer.h.23.self_attention.dense.weight": "model-00007-of-00012.safetensors", "transformer.h.23.self_attention.query_key_value.weight": "model-00007-of-00012.safetensors", "transformer.h.24.input_layernorm.bias": "model-00007-of-00012.safetensors", "transformer.h.24.input_layernorm.weight": "model-00007-of-00012.safetensors", "transformer.h.24.mlp.dense_4h_to_h.weight": "model-00008-of-00012.safetensors", "transformer.h.24.mlp.dense_h_to_4h.weight": "model-00008-of-00012.safetensors", "transformer.h.24.self_attention.dense.weight": "model-00008-of-00012.safetensors", "transformer.h.24.self_attention.query_key_value.weight": "model-00008-of-00012.safetensors", "transformer.h.3.input_layernorm.bias": "model-00008-of-00012.safetensors", "transformer.h.3.input_layernorm.weight": "model-00008-of-00012.safetensors", "transformer.h.3.mlp.dense_4h_to_h.weight": "model-00008-of-00012.safetensors", "transformer.h.3.mlp.dense_h_to_4h.weight": "model-00008-of-00012.safetensors", "transformer.h.3.self_attention.dense.weight": "model-00008-of-00012.safetensors", "transformer.h.3.self_attention.query_key_value.weight": "model-00008-of-00012.safetensors", "transformer.h.4.input_layernorm.bias": "model-00008-of-00012.safetensors", "transformer.h.4.input_layernorm.weight": "model-00008-of-00012.safetensors", "transformer.h.4.mlp.dense_4h_to_h.weight": "model-00008-of-00012.safetensors", "transformer.h.4.mlp.dense_h_to_4h.weight": "model-00008-of-00012.safetensors", "transformer.h.4.self_attention.dense.weight": "model-00009-of-00012.safetensors", "transformer.h.4.self_attention.query_key_value.weight": "model-00009-of-00012.safetensors", "transformer.h.5.input_layernorm.bias": "model-00009-of-00012.safetensors", "transformer.h.5.input_layernorm.weight": "model-00009-of-00012.safetensors", "transformer.h.5.mlp.dense_4h_to_h.weight": "model-00009-of-00012.safetensors", "transformer.h.5.mlp.dense_h_to_4h.weight": "model-00009-of-00012.safetensors", "transformer.h.5.self_attention.dense.weight": "model-00009-of-00012.safetensors", "transformer.h.5.self_attention.query_key_value.weight": "model-00009-of-00012.safetensors", "transformer.h.25.input_layernorm.bias": "model-00009-of-00012.safetensors", "transformer.h.25.input_layernorm.weight": "model-00009-of-00012.safetensors", "transformer.h.25.mlp.dense_4h_to_h.weight": "model-00009-of-00012.safetensors", "transformer.h.25.mlp.dense_h_to_4h.weight": "model-00009-of-00012.safetensors", "transformer.h.25.self_attention.dense.weight": "model-00009-of-00012.safetensors", "transformer.h.25.self_attention.query_key_value.weight": "model-00009-of-00012.safetensors", "transformer.h.26.input_layernorm.bias": "model-00009-of-00012.safetensors", "transformer.h.26.input_layernorm.weight": "model-00009-of-00012.safetensors", "transformer.h.26.mlp.dense_4h_to_h.weight": "model-00009-of-00012.safetensors", "transformer.h.26.mlp.dense_h_to_4h.weight": "model-00010-of-00012.safetensors", "transformer.h.26.self_attention.dense.weight": "model-00010-of-00012.safetensors", "transformer.h.26.self_attention.query_key_value.weight": "model-00010-of-00012.safetensors", "transformer.h.27.input_layernorm.bias": "model-00010-of-00012.safetensors", "transformer.h.27.input_layernorm.weight": "model-00010-of-00012.safetensors", "transformer.h.27.mlp.dense_4h_to_h.weight": "model-00010-of-00012.safetensors", "transformer.h.27.mlp.dense_h_to_4h.weight": "model-00010-of-00012.safetensors", "transformer.h.27.self_attention.dense.weight": "model-00010-of-00012.safetensors", "transformer.h.27.self_attention.query_key_value.weight": "model-00010-of-00012.safetensors", "transformer.h.6.input_layernorm.bias": "model-00010-of-00012.safetensors", "transformer.h.6.input_layernorm.weight": "model-00010-of-00012.safetensors", "transformer.h.6.mlp.dense_4h_to_h.weight": "model-00010-of-00012.safetensors", "transformer.h.6.mlp.dense_h_to_4h.weight": "model-00010-of-00012.safetensors", "transformer.h.6.self_attention.dense.weight": "model-00010-of-00012.safetensors", "transformer.h.6.self_attention.query_key_value.weight": "model-00010-of-00012.safetensors", "transformer.h.7.input_layernorm.bias": "model-00010-of-00012.safetensors", "transformer.h.7.input_layernorm.weight": "model-00010-of-00012.safetensors", "transformer.h.7.mlp.dense_4h_to_h.weight": "model-00011-of-00012.safetensors", "transformer.h.7.mlp.dense_h_to_4h.weight": "model-00011-of-00012.safetensors", "transformer.h.7.self_attention.dense.weight": "model-00011-of-00012.safetensors", "transformer.h.7.self_attention.query_key_value.weight": "model-00011-of-00012.safetensors", "transformer.h.8.input_layernorm.bias": "model-00011-of-00012.safetensors", "transformer.h.8.input_layernorm.weight": "model-00011-of-00012.safetensors", "transformer.h.8.mlp.dense_4h_to_h.weight": "model-00011-of-00012.safetensors", "transformer.h.8.mlp.dense_h_to_4h.weight": "model-00011-of-00012.safetensors", "transformer.h.8.self_attention.dense.weight": "model-00011-of-00012.safetensors", "transformer.h.8.self_attention.query_key_value.weight": "model-00011-of-00012.safetensors", "transformer.h.9.input_layernorm.bias": "model-00011-of-00012.safetensors", "transformer.h.9.input_layernorm.weight": "model-00011-of-00012.safetensors", "transformer.h.9.mlp.dense_4h_to_h.weight": "model-00011-of-00012.safetensors", "transformer.h.9.mlp.dense_h_to_4h.weight": "model-00011-of-00012.safetensors", "transformer.h.9.self_attention.dense.weight": "model-00012-of-00012.safetensors", "transformer.h.9.self_attention.query_key_value.weight": "model-00012-of-00012.safetensors", "transformer.ln_f.bias": "model-00012-of-00012.safetensors", "transformer.ln_f.weight": "model-00012-of-00012.safetensors", "transformer.word_embeddings.weight": "model-00012-of-00012.safetensors"}}

special_tokens_map.json

@@ -0,0 +1,36 @@
+{
+  "additional_special_tokens": [
+    ">>TITLE<<",
+    ">>ABSTRACT<<",
+    ">>INTRODUCTION<<",
+    ">>SUMMARY<<",
+    ">>COMMENT<<",
+    ">>ANSWER<<",
+    ">>QUESTION<<",
+    ">>DOMAIN<<",
+    ">>PREFIX<<",
+    ">>SUFFIX<<",
+    ">>MIDDLE<<"
+  ],
+  "bos_token": {
+    "content": ">>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,135 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": ">>TITLE<<",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": ">>ABSTRACT<<",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": ">>INTRODUCTION<<",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": ">>SUMMARY<<",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": ">>COMMENT<<",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": ">>ANSWER<<",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": ">>QUESTION<<",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": ">>DOMAIN<<",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": ">>PREFIX<<",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": ">>SUFFIX<<",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": ">>MIDDLE<<",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "500": {
+      "content": ">>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    ">>TITLE<<",
+    ">>ABSTRACT<<",
+    ">>INTRODUCTION<<",
+    ">>SUMMARY<<",
+    ">>COMMENT<<",
+    ">>ANSWER<<",
+    ">>QUESTION<<",
+    ">>DOMAIN<<",
+    ">>PREFIX<<",
+    ">>SUFFIX<<",
+    ">>MIDDLE<<"
+  ],
+  "bos_token": ">>",
+  "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ 'User: \n' + message['content'] }}\n{% elif message['role'] == 'system' %}\n{{ 'System: ' + message['content'] }}\n{% elif message['role'] == 'assistant' %}\n{{ 'Falcon:\n'  + message['content']}}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ 'Falcon:' }}\n{% endif %}\n{% endfor %}",
+  "clean_up_tokenization_spaces": true,
+  "device_map": "cuda:2",
+  "eos_token": "<|endoftext|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}