Upload MambaForCausalLM

README.md

@@ -1,9 +1,9 @@
 ---
+language:
+- it
 license: apache-2.0
 datasets:
 - DeepMount00/gquad_it
-language:
-- it
 pipeline_tag: question-answering
 ---
 
@@ -28,11 +28,11 @@ def predict(contesto, domanda):
 
     input_ids = tokenizer([prompt], return_tensors="pt").to(device)
 
-    generate_ids = model.generate(**input_ids, max_new_tokens=150, eos_token_id=8112)
+    generate_ids = model.generate(**input_ids, max_new_tokens=150, eos_token_id=0)
 
     answer = tokenizer.batch_decode(generate_ids)
     try:
-        final_answer = answer[0].split("##RISPOSTA: ")[1].split("##END")[0].strip("\n")
+        final_answer = answer[0].split("##RISPOSTA: ")[1].split('\n', 1)[0]
     except IndexError:
         final_answer = ""
     return final_answer

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "mamba_qa/checkpoint-33000",
+  "_name_or_path": "mamba_qa/checkpoint-43000",
   "architectures": [
     "MambaForCausalLM"
   ],

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c9306579a959a462a03648db26a708c2756e2eb1ef3d35684811db590bbdd06
+oid sha256:aaa63c9947dac0db02479bb3b718f2e0d8d291ef9bf1b88caeb681767d1302fc
 size 3172869936