[fix]load hadamard_block_size from config #3797
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thanks for your contribution! |
yangjianfengo1
approved these changes
Sep 2, 2025
Contributor
|
LGTM |
K11OntheBoat
reviewed
Sep 2, 2025
| self.moe_quant_type, | ||
| used_in_ep_low_latency, | ||
| estimate_total_token_nums, | ||
| getattr(layer, "hadamard_block_size", 512), |
Collaborator
There was a problem hiding this comment.
其他地方默认值是128，这里是512，需要统一一下
rsmallblue
force-pushed
the
w4a8-config
branch
from
0f92f67 to
d2abad4
Compare
rsmallblue
changed the title
rsmallblue
force-pushed
the
w4a8-config
branch
from
d2abad4 to
6131d29
Compare
rsmallblue
force-pushed
the
w4a8-config
branch
from
2c2d625 to
3cf7c70
Compare
RichardWooSJTU
approved these changes
Sep 5, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR modifies the hadamard_block_size to be loaded from config.json.
Example：
{ "architectures": [ "Ernie4_5_MoeForCausalLM" ], "bos_token_id": 1, "eos_token_id": 2, "dtype": "bfloat16", "hidden_act": "silu", "hidden_size": 8192, "intermediate_size": 28672, "max_position_embeddings": 131072, "model_type": "ernie4_5_moe", "num_attention_heads": 64, "num_key_value_heads": 8, "num_hidden_layers": 54, "pad_token_id": 0, "rms_norm_eps": 1e-05, "use_cache": false, "vocab_size": 103424, "rope_theta": 500000, "use_rmsnorm": true, "use_bias": false, "moe_num_experts": 64, "moe_layer_start_index": 3, "moe_intermediate_size": 3584, "moe_capacity": [64,64,64], "moe_gate": "topk", "moe_k": 8, "moe_layer_interval": 1, "moe_use_aux_free": true, "num_nextn_predict_layers": 1, "tie_word_embeddings": false, "is_quantized": true, "quantization_config":{ "dense_quant_type":"block_wise_fp8", "moe_quant_type":"w4a8", "quantization":"mix_quant", "is_permuted": true, "hadamard_block_size": 512 } }