Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add llama3.1 to HELM API #2835

Merged
merged 4 commits into from
Jul 23, 2024
Merged

Add llama3.1 to HELM API #2835

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
3f880c0
Add Together Llama 3.1
Jul 23, 2024
9292cd3
Add metadata
Jul 23, 2024
7e62eeb
Update tokenizer
Jul 23, 2024
fe9d381
Update tokenizer
Jul 23, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
27 changes: 27 additions & 0 deletions src/helm/config/model_deployments.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1799,6 +1799,33 @@ model_deployments:
args:
together_model: meta-llama/Meta-Llama-3-70B

- name: together/llama-3.1-8b-instruct-turbo
model_name: meta/llama-3.1-8b-instruct-turbo
tokenizer_name: meta/llama-3.1-8b
max_sequence_length: 8192
client_spec:
class_name: "helm.clients.together_client.TogetherClient"
args:
together_model: meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo

- name: together/llama-3.1-70b-instruct-turbo
model_name: meta/llama-3.1-70b-instruct-turbo
tokenizer_name: meta/llama-3.1-8b
max_sequence_length: 8192
client_spec:
class_name: "helm.clients.together_client.TogetherClient"
args:
together_model: meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo

- name: together/llama-3.1-405b-instruct-turbo
model_name: meta/llama-3.1-405b-instruct-turbo
tokenizer_name: meta/llama-3.1-8b
max_sequence_length: 4096
client_spec:
class_name: "helm.clients.together_client.TogetherClient"
args:
together_model: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo

- name: together/llama-3-8b-chat
model_name: meta/llama-3-8b-chat
tokenizer_name: meta/llama-3-8b
Expand Down
27 changes: 27 additions & 0 deletions src/helm/config/model_metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1418,6 +1418,33 @@ models:
num_parameters: 70000000000
release_date: 2024-04-18
tags: [TEXT_MODEL_TAG, LIMITED_FUNCTIONALITY_TEXT_MODEL_TAG]

- name: meta/llama-3.1-8b-instruct-turbo
display_name: Llama 3.1 (8B)
description: Llama 3.1 is a family of language models.
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lol

creator_organization_name: Meta
access: open
num_parameters: 8000000000
release_date: 2024-07-23
tags: [TEXT_MODEL_TAG, LIMITED_FUNCTIONALITY_TEXT_MODEL_TAG]

- name: meta/llama-3.1-70b-instruct-turbo
display_name: Llama 3.1 (70B)
description: Llama 3.1 is a family of language models.
creator_organization_name: Meta
access: open
num_parameters: 70000000000
release_date: 2024-07-23
tags: [TEXT_MODEL_TAG, LIMITED_FUNCTIONALITY_TEXT_MODEL_TAG]

- name: meta/llama-3.1-405b-instruct-turbo
display_name: Llama 3.1 (405B)
description: Llama 3.1 is a family of language models.
creator_organization_name: Meta
access: open
num_parameters: 405000000000
release_date: 2024-07-23
tags: [TEXT_MODEL_TAG, LIMITED_FUNCTIONALITY_TEXT_MODEL_TAG]

- name: meta/llama-3-8b-chat
display_name: Llama 3 Instruct (8B)
Expand Down
8 changes: 8 additions & 0 deletions src/helm/config/tokenizer_configs.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -303,6 +303,14 @@ tokenizer_configs:
prefix_token: "<|begin_of_text|>"
end_of_text_token: "<|end_of_text|>"

- name: meta/llama-3.1-8b
tokenizer_spec:
class_name: "helm.tokenizers.huggingface_tokenizer.HuggingFaceTokenizer"
args:
pretrained_model_name_or_path: meta-llama/Meta-Llama-3.1-8B-Instruct
prefix_token: "<|begin_of_text|>"
end_of_text_token: "<|end_of_text|>"

# 01-ai
- name: 01-ai/Yi-6B
tokenizer_spec:
Expand Down