From e184b06c4b44d4635d6745603a6538f3a8384b65 Mon Sep 17 00:00:00 2001 From: Daniel Han-Chen Date: Tue, 23 Jan 2024 19:28:17 +1100 Subject: [PATCH] Update mistral.py --- unsloth/models/mistral.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/unsloth/models/mistral.py b/unsloth/models/mistral.py index 06a3660d45..7fb002fa1f 100644 --- a/unsloth/models/mistral.py +++ b/unsloth/models/mistral.py @@ -69,6 +69,8 @@ def MistralAttention_fast_forward( Q = Q.view(bsz, q_len, n_heads, head_dim).transpose(1, 2) K = K.view(bsz, q_len, n_kv_heads, head_dim).transpose(1, 2) V = V.view(bsz, q_len, n_kv_heads, head_dim).transpose(1, 2) + print(hidden_states.shape) + print(Q.shape, K.shape, V.shape) kv_seq_len = K.shape[-2] if past_key_value is not None: