opt : fix n_outputs

ggml-ci

Author: ggerganov

src/llama-context.cpp

@@ -1955,6 +1955,17 @@ void llama_context::opt_epoch_iter(
             //}
             llama_ubatch ubatch = kv_self->ubatch_next(sbatch, cparams.n_ubatch, embd_pooled);
 
+            n_outputs = ubatch.n_tokens;
+
+            printf("ubatch.n_tokens = %d\n", ubatch.n_tokens);
+
+            // TODO: not sure if this is needed
+            if (!kv_self->find_slot(ubatch)) {
+                LLAMA_LOG_WARN("%s: failed to find KV cache slot for ubatch of size %d\n", __func__, ubatch.n_tokens);
+
+                GGML_ABORT("TODO: handle this error");
+            }
+
             auto * gf = graph_init();
             auto res = graph_build(ctx_compute.get(), gf, ubatch, LLM_GRAPH_TYPE_DEFAULT);
 
@@ -1970,7 +1981,7 @@ void llama_context::opt_epoch_iter(
                 };
                 ctx_compute_opt = ggml_init(params);
             }
-            ggml_opt_prepare_alloc(opt_ctx, ctx_compute_opt, gf, res->get_tokens(), ggml_graph_node(gf, -1));
+            ggml_opt_prepare_alloc(opt_ctx, ctx_compute_opt, gf, res->get_tokens(), res->get_logits());
             ggml_opt_alloc(opt_ctx, train);
             //llama_set_inputs(*lctx, ubatch);
             res->set_inputs(&ubatch);