chaski-7B

README.md

@@ -32,14 +32,14 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1e-06
+- learning_rate: 0.0002
 - train_batch_size: 1
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: constant
 - lr_scheduler_warmup_ratio: 0.03
-- num_epochs: 3
+- num_epochs: 5
 
 ### Training results
 

adapter_config.json

@@ -19,13 +19,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "v_proj",
+    "k_proj",
+    "down_proj",
     "gate_proj",
     "up_proj",
-    "down_proj",
+    "v_proj",
     "q_proj",
-    "k_proj"
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_rslora": false

generation_config.json

@@ -1,7 +1,6 @@
 {
   "_from_model_config": true,
-  "bos_token_id": 32001,
-  "eos_token_id": 32000,
-  "pad_token_id": 32000,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
   "transformers_version": "4.36.2"
 }

runs/Mar20_12-37-02_lab/events.out.tfevents.1710938223.lab.32302.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8499d8c373c99b6c8cccac7415ce14819b00c4e0ebabff82f42d1360e3bbc83
+size 4733

runs/Mar20_12-38-11_lab/events.out.tfevents.1710938291.lab.32586.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:392edf39fc15b2160595c089bff4af3806647d1e9c2ede3a42091ce3ac6e3b68
+size 5543

special_tokens_map.json

@@ -1,23 +1,25 @@
 {
-  "additional_special_tokens": [
-    {
-      "content": "<|im_end|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "<|im_start|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    }
-  ],
-  "bos_token": "<|im_start|>",
-  "eos_token": "<|im_end|>",
-  "pad_token": "<|im_end|>",
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": true,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
   "unk_token": {
     "content": "<unk>",
     "lstrip": true,

tokenizer.json

@@ -50,8 +50,8 @@
       "id": 32001,
       "content": "<|im_start|>",
       "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
+      "lstrip": true,
+      "rstrip": true,
       "normalized": false,
       "special": true
     }

tokenizer_config.json

@@ -36,21 +36,18 @@
     },
     "32001": {
       "content": "<|im_start|>",
-      "lstrip": false,
+      "lstrip": true,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     }
   },
-  "additional_special_tokens": [
-    "<|im_end|>",
-    "<|im_start|>"
-  ],
-  "bos_token": "<|im_start|>",
-  "chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
+  "chat_template": "{{ bos_token }}{% for message in messages %}<|im_start|> {{ message.role }}\n{{ message.content }}{% if message.role == 'assistant' %}{{ eos_token }}{% endif %}",
   "clean_up_tokenization_spaces": false,
-  "eos_token": "<|im_end|>",
+  "eos_token": "</s>",
   "legacy": true,
   "max_length": null,
   "model_max_length": 1000000000000000019884624838656,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:990aae14fbee58b82bc9875a5070f80d8226c4bdaed43fd024ecca7d1820e735
+oid sha256:ad6ea7e984b94d60cd1ce82807509a4605d551c369dae971949ade7690f31b5b
 size 4664