Amazon Bedrock's Claude warns "Token indices sequence length is longer than the specified maximum sequence length for this model (6526 > 1024)

Hi, I just tested it, and it worked well in my enviroment.

First, I assume you have update the repo to the latest version, of which the commit hash should be cfecb6215c8171ba2580949a426193ccaf4c5b7f by the time I write this answer.

Then, edit this section below in the file 02_Summarization/02.long-text-summarization-titan.ipynb,

from langchain.llms.bedrock import Bedrock

# llm = Bedrock(
#     model_id="amazon.titan-tg1-large",
#     model_kwargs={
#         "maxTokenCount": 4096,
#         "stopSequences": [],
#         "temperature": 0,
#         "topP": 1,
#     },
#     client=boto3_bedrock,
# )
llm = Bedrock(
    # model_id="anthropic.claude-v1",
    model_id="anthropic.claude-v2",
    model_kwargs={
        "max_tokens_to_sample":4096,
                 "temperature":0.5,
                 "top_k":250,
                 "top_p":0.5,
                 "stop_sequences":[]
    },
    client=boto3_bedrock,
)
shareholder_letter = "./letters/2022-letter.txt"

with open(shareholder_letter, "r") as file:
    letter = file.read()
    
llm.get_num_tokens(letter)

Then run 02_Summarization/02.long-text-summarization-titan.ipynb from the beginning, you should get the output: 6526