Upload MemoryLLM

README.md

@@ -40,15 +40,8 @@ messages = [{
 inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True)
 inputs = inputs[:, 1:] # remove bos token
 
-terminators = [
-    tokenizer.eos_token_id,
-    tokenizer.convert_tokens_to_ids("<|eot_id|>")
-]
-
-outputs = model.generate(input_ids=inputs.cuda()[:, 1:],
-                         max_new_tokens=20,
-                         eos_token_id=terminators)
-
+outputs = model.generate(input_ids=inputs.cuda(),
+                         max_new_tokens=20)
 response = tokenizer.decode(outputs[0])
 
 outputs = model.generate(inputs=input_ids.cuda(), attention_mask=attention_mask.cuda(), max_new_tokens=10)

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "memoryllm-8b-chat",
+  "_name_or_path": "locals/memoryllm-8b-chat",
   "add_bos_embedding": true,
   "add_decoder_lora": true,
   "architectures": [
@@ -25,7 +25,8 @@
       "k_proj",
       "up_proj",
       "down_proj",
-      "gate_proj"
+      "gate_proj",
+      "lm_head"
     ]
   },
   "max_length": 512,
@@ -40,7 +41,7 @@
   "num_key_value_heads": 8,
   "num_memory_tokens": 12800,
   "num_tokens": 256,
-  "pad_token_id": 128001,
+  "pad_token_id": 128009,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
   "rope_scaling": {
@@ -49,9 +50,15 @@
   },
   "rope_theta": 500000.0,
   "shrink_to_one_embedding": true,
+  "special_token_ids": [
+    128006,
+    128007,
+    128009
+  ],
   "tie_word_embeddings": false,
   "torch_dtype": "float32",
   "transformers_version": "4.43.0.dev0",
+  "tune_special_tokens": true,
   "use_cache": true,
   "vocab_size": 128256
 }

model-00001-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2713410b4c31a0923361c8e21553a13a43c5af08aacca91305d8bfd555f86730
+oid sha256:7af65cce95946eeee8693d00b8a1d6695de24f3801256061f769082ec3bebe2f
 size 6710886520

model-00002-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3d5edfd6ebf9e3bcadf6143f8d7924f87409842b9210a00b3a92615f9ee1b7fa
+oid sha256:c3ec2e5cccec05a70c3a968b4a3c8a0f05f49aa028aad8218115e2b4100e90b6
 size 4917689241

model-00003-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0dae25919a3894e7ecf58331c96d83ba1b6808337de473bc2e7bd1d98c863cb
+oid sha256:116b96cb823882cfe86460aa84c99ca793deafafe8c40590f37e7e0a8412be45
 size 4883929088

model-00004-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:13b18147bef5aa85bc9bcc820893862c9803d1da156b71019c19b3bd2e97df57
+oid sha256:3e17b030cf85a2824f567c948f3e269a836092307c9dde1857063fdc3891b872
 size 4816854808

model-00005-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:58733b1be1d8b93ec09c833ebd28e70a9e4ef3f2e742215b7e68cf5edfab294a
+oid sha256:e32e3065fa8d4716745ea8a5427ba20467f591fd06df27d120a93a1e41223805
 size 4883929264

model-00006-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50b8f69fc77bb9b0dee767b56100d10bacea05882cb15e4aa92edf9949034a86
+oid sha256:8efe66af244af0fadf5a3ab3dd080f2ff7311a65653179251c9f5ee204b61ba4
 size 4816854856

model-00007-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:410de571515c3fd957296da3d6a1e1579820f575fef227ce5517496f5d4e162c
+oid sha256:e44fc390f8a968a6534b3369b5d4387a60cad4204c5a6a5e9ac0b46c7c19a308
 size 4883929264

model-00008-of-00008.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a90c176b152a64e78d5b756b9562ac1a251009ce70bb766b3de43f5232aad9f
+oid sha256:dcf4b278965fd94af150585d96f3051c1c1dada3b54246757b78dbf6ad8d58b3
 size 3220526344