Need to update params since tensor changed.

integration-tests/models/__snapshots__/test_flash_awq/test_flash_llama_awq_all_params.json

@@ -39,61 +39,60 @@
         "text": "\n"
       },
       {
-        "id": 4013,
+        "id": 1576,
-        "logprob": -2.203125,
+        "logprob": -1.2138672,
         "special": false,
-        "text": "This"
+        "text": "The"
       },
       {
-        "id": 1139,
+        "id": 3611,
-        "logprob": -0.23693848,
+        "logprob": -0.9482422,
         "special": false,
-        "text": " question"
+        "text": " title"
       },
       {
-        "id": 756,
+        "id": 310,
         "logprob": 0.0,
         "special": false,
-        "text": " has"
+        "text": " of"
       },
       {
-        "id": 1063,
+        "id": 445,
-        "logprob": -0.076538086,
+        "logprob": -0.140625,
         "special": false,
-        "text": " been"
+        "text": " this"
       },
       {
-        "id": 4433,
+        "id": 1400,
+        "logprob": -0.2286377,
+        "special": false,
+        "text": " post"
+      },
+      {
+        "id": 1795,
+        "logprob": -0.5253906,
+        "special": false,
+        "text": " might"
+      },
+      {
+        "id": 505,
+        "logprob": -0.30444336,
+        "special": false,
+        "text": " have"
+      },
+      {
+        "id": 366,
         "logprob": 0.0,
         "special": false,
-        "text": " asked"
+        "text": " you"
       },
       {
-        "id": 1784,
+        "id": 7291,
-        "logprob": -1.1367188,
+        "logprob": -0.68652344,
         "special": false,
-        "text": " many"
+        "text": " thinking"
-      },
-      {
-        "id": 3064,
-        "logprob": 0.0,
-        "special": false,
-        "text": " times"
-      },
-      {
-        "id": 322,
-        "logprob": -1.7460938,
-        "special": false,
-        "text": " and"
-      },
-      {
-        "id": 306,
-        "logprob": 0.0,
-        "special": false,
-        "text": " I"
       }
-    ],
+    ]
-    "top_tokens": null
   },
-  "generated_text": "What is Deep Learning?\nThis question has been asked many times and I"
+  "generated_text": "What is Deep Learning?\nThe title of this post might have you thinking"
 }

integration-tests/models/__snapshots__/test_flash_falcon/test_flash_falcon_all_params.json

@@ -11,22 +11,22 @@
       },
       {
         "id": 1622,
-        "logprob": -7.8125,
+        "logprob": -7.7421875,
         "text": "af"
       },
       {
         "id": 249,
-        "logprob": -4.5,
+        "logprob": -4.484375,
         "text": "at"
       },
       {
         "id": 1480,
-        "logprob": -10.875,
+        "logprob": -10.890625,
         "text": "ron"
       },
       {
         "id": 37,
-        "logprob": -3.6875,
+        "logprob": -3.6757812,
         "text": ":"
       }
     ],
@@ -34,65 +34,65 @@
     "tokens": [
       {
         "id": 836,
-        "logprob": -1.265625,
+        "logprob": -1.2822266,
         "special": false,
         "text": " i"
       },
       {
         "id": 18,
-        "logprob": -0.119628906,
+        "logprob": -0.14953613,
         "special": false,
         "text": "'"
       },
       {
-        "id": 298,
+        "id": 88,
-        "logprob": -2.265625,
+        "logprob": -0.11138916,
         "special": false,
-        "text": "ve"
+        "text": "m"
       },
       {
-        "id": 650,
+        "id": 416,
-        "logprob": -0.49804688,
+        "logprob": -0.074401855,
         "special": false,
-        "text": " been"
+        "text": " not"
       },
       {
-        "id": 1241,
+        "id": 830,
-        "logprob": 0.0,
+        "logprob": -2.2148438,
         "special": false,
-        "text": " using"
+        "text": " even"
+      },
+      {
+        "id": 1310,
+        "logprob": -0.19958496,
+        "special": false,
+        "text": " sure"
+      },
+      {
+        "id": 637,
+        "logprob": -0.19787598,
+        "special": false,
+        "text": " what"
       },
       {
         "id": 334,
-        "logprob": 0.0,
+        "logprob": -0.6386719,
         "special": false,
         "text": " it"
       },
       {
-        "id": 312,
+        "id": 304,
-        "logprob": -1.2421875,
+        "logprob": 0.0,
         "special": false,
-        "text": " for"
+        "text": " is"
-      },
-      {
-        "id": 909,
-        "logprob": -0.99609375,
-        "special": false,
-        "text": " years"
       },
       {
         "id": 193,
-        "logprob": -0.30273438,
+        "logprob": 0.0,
         "special": false,
         "text": "\n"
-      },
-      {
-        "id": 807,
-        "logprob": -1.078125,
-        "special": false,
-        "text": "ik"
       }
     ]
   },
-  "generated_text": "Girafatron is obsessed with giraffes, the most glorious animal on the face of this Earth. Giraftron believes all other animals are irrelevant when compared to the glorious majesty of the giraffe.\nDaniel: Hello, Girafatron!\nGirafatron: i've been using it for years\nik"
+  "generated_text": "Girafatron is obsessed with giraffes, the most glorious animal on the face of this Earth. Giraftron believes all other animals are irrelevant when compared to the glorious majesty of the giraffe.\nDaniel: Hello, Girafatron!\nGirafatron: i'm not even sure what it is\n"
 }

integration-tests/models/__snapshots__/test_flash_llama/test_flash_llama_all_params.json

@@ -1,8 +1,8 @@
 {
   "details": {
     "best_of_sequences": null,
-    "finish_reason": "stop_sequence",
+    "finish_reason": "length",
-    "generated_tokens": 5,
+    "generated_tokens": 10,
     "prefill": [
       {
         "id": 1,
@@ -28,32 +28,61 @@
         "special": false,
         "text": " failed"
       },
+      {
+        "id": 363,
+        "logprob": -1.1064453,
+        "special": false,
+        "text": " for"
+      },
+      {
+        "id": 278,
+        "logprob": -0.7709961,
+        "special": false,
+        "text": " the"
+      },
+      {
+        "id": 1494,
+        "logprob": -0.13464355,
+        "special": false,
+        "text": " following"
+      },
+      {
+        "id": 9590,
+        "logprob": -0.1986084,
+        "special": false,
+        "text": " reasons"
+      },
       {
         "id": 29901,
-        "logprob": -0.44970703,
+        "logprob": 0.0,
         "special": false,
         "text": ":"
       },
       {
-        "id": 4829,
+        "id": 13,
-        "logprob": -1.8339844,
-        "special": false,
-        "text": " Error"
-      },
-      {
-        "id": 297,
-        "logprob": -1.0556641,
-        "special": false,
-        "text": " in"
-      },
-      {
-        "id": 1243,
         "logprob": 0.0,
         "special": false,
-        "text": " test"
+        "text": "\n"
+      },
+      {
+        "id": 1576,
+        "logprob": -0.6166992,
+        "special": false,
+        "text": "The"
+      },
+      {
+        "id": 13877,
+        "logprob": -0.91845703,
+        "special": false,
+        "text": " requested"
+      },
+      {
+        "id": 3988,
+        "logprob": -0.7680664,
+        "special": false,
+        "text": " URL"
       }
-    ],
+    ]
-    "top_tokens": null
   },
-  "generated_text": "Test request failed: Error in test"
+  "generated_text": "Test request failed for the following reasons:\nThe requested URL"
 }

server/text_generation_server/models/seq2seq_lm.py

@@ -736,7 +736,7 @@ class Seq2SeqLM(Model):
                         [self.tokenizer.bos_token_id],
                         [float("nan")],
                         [self.tokenizer.bos_token],
-                        [false]
+                        [False]
                     )
                 else:
                     prefill_tokens = None