Fix BOS (should be added by the tokenizer). Fix EOS (in config.json -> must be <|eot_id|>). Remove spaces that were aded after special tags (so that round-trip tests can pass and some bugs avoided when inferring with the model). More complete generation_config.json

Lucie-7B-q4_k_m.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32ad1ff565d483ec8b6a1d16ad1b64866f7db1203568daebeb5f4d9b0d78682f
-size 4068731808
+oid sha256:d41d59729e4e0e21d6bbb41b6d146643cbe46cd49d849fbef8eb9d318255f883
+size 4068731936

config.json

@@ -1,82 +1,31 @@
 {
-  "vocab_size": 65024,
-  "max_position_embeddings": 4096,
+  "_name_or_path": "/data-server/models/text/llm/multilang/Lucie-7B/instruct/llama_instruct/recipe2_fullpipe/base",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "eos_token_id": 267,
+  "head_dim": 128,
+  "hidden_act": "silu",
   "hidden_size": 4096,
+  "initializer_range": 0.02,
   "intermediate_size": 12288,
-  "num_hidden_layers": 32,
+  "max_position_embeddings": 32000,
+  "mlp_bias": false,
+  "model_type": "llama",
   "num_attention_heads": 32,
+  "num_hidden_layers": 32,
   "num_key_value_heads": 8,
-  "hidden_act": "silu",
-  "initializer_range": 0.02,
-  "rms_norm_eps": 1e-05,
+  "pad_token_id": 3,
   "pretraining_tp": 1,
-  "use_cache": true,
-  "rope_theta": 20000000,
+  "rms_norm_eps": 1e-05,
   "rope_scaling": null,
-  "attention_bias": false,
-  "attention_dropout": 0.0,
-  "return_dict": true,
-  "output_hidden_states": false,
-  "output_attentions": false,
-  "torchscript": false,
-  "torch_dtype": "bfloat16",
-  "use_bfloat16": false,
-  "tf_legacy_loss": false,
-  "pruned_heads": {},
+  "rope_theta": 20000000.0,
   "tie_word_embeddings": false,
-  "is_encoder_decoder": false,
-  "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,
-  "chunk_size_feed_forward": 0,
-  "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": [
-    "LlamaForCausalLM"
-  ],
-  "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": 0,
-  "pad_token_id": 3,
-  "eos_token_id": 1,
-  "sep_token_id": null,
-  "decoder_start_token_id": null,
-  "task_specific_params": null,
-  "problem_type": null,
-  "_name_or_path": "",
-  "transformers_version": "4.36.1",
-  "model_type": "llama"
-}
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.3",
+  "use_cache": true,
+  "vocab_size": 65024
+}

generation_config.json

@@ -1,7 +1,9 @@
 {
-  "_from_model_config": true,
   "bos_token_id": 0,
   "eos_token_id": 267,
   "pad_token_id": 3,
+  "max_length": 32000,
+  "do_sample": true,
+  "temperature": 0.6,
   "transformers_version": "4.46.3"
-}
+}

tokenizer.json

@@ -289,6 +289,27 @@
       {
         "type": "Fuse"
       },
+      {
+        "type": "Replace",
+        "pattern": {
+          "String": "<|start_header_id|> "
+        },
+        "content": "<|start_header_id|>"
+      },
+      {
+        "type": "Replace",
+        "pattern": {
+          "String": "<|end_header_id|> "
+        },
+        "content": "<|end_header_id|>"
+      },
+      {
+        "type": "Replace",
+        "pattern": {
+          "String": "<|eot_id|> "
+        },
+        "content": "<|eot_id|>"
+      },
       {
         "type": "Replace",
         "pattern": {

tokenizer_config.json

@@ -1,5 +1,5 @@
 {
-  "add_bos_token": false,
+  "add_bos_token": true,
   "add_eos_token": false,
   "added_tokens_decoder": {
     "0": {